├── .gitignore ├── README.md ├── lecture01 ├── bibliography.bib ├── fields.png ├── lecture01.pdf ├── lecture01.tex ├── speed.png └── style.sty ├── lecture02 ├── EyeDiagram.png ├── bibliography.bib ├── color1.png ├── eyediagram.gif ├── hsv1.png ├── lecture02.pdf ├── lecture02.tex ├── rgb1.png ├── style.sty ├── wb.png └── xyz1.png ├── lecture03 ├── bibliography.bib ├── lecture03.pdf ├── lecture03.tex └── style.sty ├── lecture04 ├── bibliography.bib ├── histogram.png ├── impulse_response.jpeg ├── input_signal_.jpeg ├── lecture04.pdf ├── lecture04.tex ├── res.png └── style.sty ├── lecture05 ├── bibliography.bib ├── biederman_outlines.png ├── fei-fei_detection.png ├── fei-fei_images.png ├── hubel_wiesel_cat.jpg ├── lazebnik.png ├── lazebnik_tiger.png ├── lecture05.pdf ├── lecture05.tex ├── seitz_gradient_direction.png ├── seitz_noise.png ├── seitz_smoothing.png └── style.sty ├── lecture06 ├── bibliography.bib ├── hough_transform.png ├── hough_transform2.png ├── hough_transform3.png ├── hough_transform4.png ├── lecture06.pdf ├── lecture06.tex ├── line_drawings.png ├── overall.png ├── ransac1.png ├── ransac2.png ├── ransac3.png ├── ransac4.png └── style.sty ├── lecture07 ├── bibliography.bib ├── eigenvalues_harris.png ├── harris_response.png ├── lecture07.pdf ├── lecture07.tex ├── local_feature_ex.png ├── scaleinvariant.png ├── sift_scale_invariant.jpg └── style.sty ├── lecture08 ├── bibliography.bib ├── bike.png ├── cmu_composite_panorama.png ├── corners.png ├── dog_pyramid.jpg ├── histogram.png ├── histogramsensitivity.png ├── homography_prob.png ├── lecture08.pdf ├── lecture08.tex ├── max.png ├── maxima.png ├── style.sty ├── transMatrices.png └── transforms.png ├── lecture09 ├── Function.png ├── Naive_Castle_Resizing.jpg ├── Optimal_Seam.png ├── Pixel_Removal.png ├── Smart_Resizing.jpg ├── Tracking.png ├── backtrack.png ├── bibliography.bib ├── camels.png ├── content_aware_resizing.png ├── energy.png ├── findseam1.png ├── forward_energy.PNG ├── forward_energy_calculation.JPG ├── lecture09.pdf ├── lecture09.tex ├── limitations_close-up.JPG ├── lion.jpg ├── main.synctex(busy) ├── muller.jpg ├── object_removal.JPG ├── objects.png ├── people.png ├── resizing_representation.JPG ├── seam_carving_limitations.JPG ├── style.sty ├── superpixels.png ├── user_constraints.PNG └── video_retargeting.JPG ├── lecture10 ├── average.png ├── bibliography.bib ├── clustering_for_summarization.png ├── complete.png ├── continuity-occlusion-2.png ├── continuity-occlusion.png ├── dendrogram2.png ├── figure-ground.png ├── gestalt-factors.png ├── k-means-algorithm.png ├── k-means-example.png ├── lecture10.pdf ├── lecture10.tex ├── mean-shift.png ├── panda.jpg ├── single.png ├── style.sty ├── tiger-segmented.png └── tiger.png ├── lecture11 ├── Bias-VarianceTrade-Off.png ├── ChoosingValueOfK.png ├── DimensionalityIssue.png ├── EuclideanIssue.png ├── basin1.png ├── basin2.png ├── bibliography.bib ├── decision_boudnaries.jpg ├── knearestneighbors.png ├── lecture11.pdf ├── lecture11.tex └── style.sty ├── lecture12 ├── Dimension_reduction_High_Level.png ├── PCA_dim.png ├── Screen_Shot_2017-11-08_at_8_29_32_PM.png ├── Usig.PNG ├── bibliography.bib ├── compare.png ├── full_pca.PNG ├── lecture12.pdf ├── lecture12.tex ├── style.sty ├── svd_first10.png ├── svd_image_compress.png ├── svd_original.png └── usigv_example.PNG ├── lecture13 ├── between-within-scatters.png ├── disgust.png ├── eigenvalues.png ├── error.png ├── error_rate.png ├── face_non-face_plot.png ├── generic.png ├── happy.png ├── lecture13.pdf ├── lecture13.synctex.gz ├── lecture13.tex ├── pca-vs-lda.png ├── reconstructed.png ├── sample.bib ├── style.sty └── variation_in_expression.png ├── lecture14 ├── Bayes_Histogram.png ├── BoW_pyramid.PNG ├── bibliography.bib ├── bow.png ├── grid_features.png ├── hist.png ├── lecture14.pdf ├── lecture14.tex ├── pyramid.PNG ├── pyramid_data1.PNG ├── pyramid_data2.PNG ├── style.sty └── visual_vocab.png ├── lecture15 ├── average_face_template.png ├── bibliography.bib ├── car_model.png ├── chart.jpg ├── coco.png ├── deformable_bike_model.png ├── deformable_head_filter.png ├── deformable_model.png ├── deformable_score1.png ├── false_neg.jpg ├── false_pos.jpg ├── feature_pyramid.jpg ├── fullyconnected.png ├── ilsvr.png ├── lecture15.pdf ├── lecture15.tex ├── object-detector-intro.png ├── person_star.png ├── person_template.png ├── pipeline.png ├── pipeline2a.png ├── pipeline2b.png ├── pipelinescoreplot.png ├── prec_recall_pic.jpg ├── precision_recall.jpg ├── predict_truth.jpg ├── results-tpfp.png ├── sliding_window_a.jpg ├── sliding_window_b.jpg ├── sliding_window_c.jpg ├── sliding_window_d.jpg ├── small_sliding_window_a.jpg ├── small_sliding_window_b.jpg ├── star.png ├── style.sty └── true_pos.jpg ├── lecture16 ├── bo.png ├── bubblestudy.png ├── confusionmatrix.png ├── fbyoutubedata.png ├── heatmap.png ├── internettraffic.png ├── lecture16.pdf ├── main.tex ├── obamacreditcard.png ├── pascal.bib └── style.sty ├── lecture17 ├── aperture-problem.jpg ├── bibliography.bib ├── grouping.png ├── lecture17.pdf ├── lecture17.tex ├── lk-pyr.png ├── lk_eigen.png ├── lk_regions.png ├── lucas-kanade-meme.jpg ├── michaelblack.png ├── quadratic.png ├── style.sty ├── treeLayer.png ├── treeOriginal.png ├── w_pyr.png └── wo_pyr.png ├── lecture18 ├── bibliography.bib ├── feature_track_example.PNG ├── fish.png ├── lecture18.pdf ├── lecture18.tex ├── man_track.png ├── style.sty ├── template.tex ├── trans_types.png └── translation.png ├── lecture19 ├── Activation_map.png ├── CNN_stacking.png ├── Conventional_schematic.png ├── DL_schematic.png ├── Hierarchical_features.png ├── Inception.png ├── Linear_regression.png ├── activation_functions.png ├── activation_functions2.PNG ├── bibliography.bib ├── densecap.png ├── format.tex ├── lecture19.pdf ├── lecture19.tex ├── loss_functions.PNG ├── neural_network.png ├── poseest.png ├── segmentation.png └── style.sty ├── lecture19_new ├── activation_functions.png ├── backprop_pic_1.png ├── backprop_pic_2.png ├── gradient_descent_code.png ├── hello.txt ├── lecture19.pdf ├── lecture19.tex ├── linearly_separable.png ├── neural_network.png ├── style.sty └── weights.png ├── lecture20 ├── Capture.PNG ├── Inception.png ├── Lecture 20.pdf ├── main.tex ├── maxpooling.png ├── relu.png ├── structure.png └── style.sty └── template ├── bibliography.bib ├── format.tex ├── style.sty └── template.tex /.gitignore: -------------------------------------------------------------------------------- 1 | *.log 2 | *.out 3 | *.bbl 4 | *.bgl 5 | *.DS_Store -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # CS131 Computer Vision: Foundations and Applications 2 | 3 | ## Instructions for creating class notes 4 | 5 | ### Overall workflow 6 | 1. Fork this repository by clicking the "Fork" button on the top right of [this 7 | page](https://github.com/StanfordVL/cs131_notes). 8 | 9 | 2. Next, clone your forked repository into your local machine: 10 | ``` 11 | git clone github.com/YOUR_GITHUB_ACCOUNT/cs131_notes.git 12 | ``` 13 | 3. Give others permission to commit to your forked repository by clicking on 14 | "Settings" and then "Collaborators". 15 | 3. Write up the class notes. 16 | 4. Push your changes to your forked repository. 17 | 5. Send a pull request to the official 18 | [repository](https://github.com/StanfordVL/cs131_notes). 19 | 20 | ### Downloading the software and testing the repository. 21 | 1. Download [pdfltex](https://www.tug.org/applications/pdftex/) so that you can 22 | compile the tex documents. You can also use [Overleaf](overleaf.com) or 23 | [Sharelatex](sharelatex.com) to compile your tex documents. 24 | 25 | 2. Go into the template folder and compile the format and template files: 26 | ``` 27 | cd template 28 | 29 | pdflatex format.tex 30 | bibtex format 31 | pdflatex format.tex 32 | 33 | pdflatex template.tex 34 | bibtex template 35 | pdflatex template.tex 36 | ``` 37 | 38 | 3. Make sure that template.pdf and format.pdf have been generated and are correctly 39 | displayed. Read format.pdf to understand how we expect the class notes to be 40 | formatted. 41 | 42 | ### Writing class notes for a given lecture 43 | 1. Create a folder called `LectureXX` where `XX` is `01` for the first lecture 44 | or `18` for the eighteenth lecture. 45 | 46 | 2. Copy over template.tex and bibliography.bib to your folder. 47 | ``` 48 | cp template/template.tex lectureXX/lectureXX.tex 49 | cp template/bibliography.bib lectureXX/bibliography.bib 50 | ``` 51 | 52 | 3. Write the lecture notes. Make sure to include pictures, references and 53 | tables. If you add images, make sure to give credit to the source of those 54 | images. 55 | 56 | 4. Compile your lecture template to make sure that it formats correctly. 57 | ``` 58 | cd lectureXX 59 | pdflatex lectureXX.tex 60 | bibtex lectureXX 61 | pdflatex lectureXX.tex 62 | ``` 63 | 64 | 5. Submit a pull request to have your lecture notes merged. 65 | -------------------------------------------------------------------------------- /lecture01/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{krishna2017visual, 2 | title={Visual genome: Connecting language and vision using crowdsourced dense image annotations}, 3 | author={Krishna, Ranjay and Zhu, Yuke and Groth, Oliver and Johnson, Justin and Hata, Kenji and Kravitz, Joshua and Chen, Stephanie and Kalantidis, Yannis and Li, Li-Jia and Shamma, David A and others}, 4 | journal={International Journal of Computer Vision}, 5 | volume={123}, 6 | number={1}, 7 | pages={32--73}, 8 | year={2017}, 9 | publisher={Springer} 10 | } 11 | 12 | @article{papert1966summer, 13 | title={The summer vision project}, 14 | author={Papert, Seymour A}, 15 | year={1966} 16 | } 17 | 18 | @article{hubel, 19 | title={Receptive fields, binocular interaction and functional architecture in the cat's visual cortex}, 20 | author={Hubel, David H and Wiesel, Torsten N}, 21 | journal={The Journal of physiology}, 22 | volume={160}, 23 | number={1}, 24 | pages={106--154}, 25 | year={1962}, 26 | publisher={Wiley Online Library} 27 | } 28 | 29 | @article{speed, 30 | title={Speed of processing in the human visual system}, 31 | author={Thorpe, Simon and Fize, Denise and Marlot, Catherine}, 32 | journal={nature}, 33 | volume={381}, 34 | number={6582}, 35 | pages={520}, 36 | year={1996}, 37 | publisher={Nature Publishing Group} 38 | } 39 | 40 | @article{failure, 41 | title={On the failure to detect changes in scenes across brief interruptions}, 42 | author={Rensink, Ronald A and Kevin O'Regan, J and Clark, James J}, 43 | journal={Visual cognition}, 44 | volume={7}, 45 | number={1-3}, 46 | pages={127--145}, 47 | year={2000}, 48 | publisher={Taylor \& Francis} 49 | } 50 | 51 | @article{multipleview, 52 | title={Multiple view geometry}, 53 | author={Heyden, Anders and Pollefeys, Marc}, 54 | journal={Emerging topics in computer vision}, 55 | pages={45--107}, 56 | year={2005}, 57 | publisher={Prentice Hall PTR} 58 | } 59 | 60 | @inproceedings{multicommunity, 61 | title={Multi-view stereo for community photo collections}, 62 | author={Goesele, Michael and Snavely, Noah and Curless, Brian and Hoppe, Hugues and Seitz, Steven M}, 63 | booktitle={Computer Vision, 2007. ICCV 2007. IEEE 11th International Conference on}, 64 | pages={1--8}, 65 | year={2007}, 66 | organization={IEEE} 67 | } 68 | 69 | @article{grasping, 70 | title={Robotic grasping of novel objects using vision}, 71 | author={Saxena, Ashutosh and Driemeyer, Justin and Ng, Andrew Y}, 72 | journal={The International Journal of Robotics Research}, 73 | volume={27}, 74 | number={2}, 75 | pages={157--173}, 76 | year={2008}, 77 | publisher={Sage Publications Sage UK: London, England} 78 | } -------------------------------------------------------------------------------- /lecture01/fields.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture01/fields.png -------------------------------------------------------------------------------- /lecture01/lecture01.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture01/lecture01.pdf -------------------------------------------------------------------------------- /lecture01/speed.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture01/speed.png -------------------------------------------------------------------------------- /lecture01/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput 302 | -------------------------------------------------------------------------------- /lecture02/EyeDiagram.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture02/EyeDiagram.png -------------------------------------------------------------------------------- /lecture02/bibliography.bib: -------------------------------------------------------------------------------- 1 | @book{palmer1999vision, 2 | title={Vision science: Photons to phenomenology}, 3 | author={Palmer, Stephen E}, 4 | year={1999}, 5 | publisher={MIT press} 6 | } 7 | 8 | @book{wandell1995foundations, 9 | title={Foundations of vision.}, 10 | author={Wandell, Brian A}, 11 | year={1995}, 12 | publisher={Sinauer Associates} 13 | } 14 | -------------------------------------------------------------------------------- /lecture02/color1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture02/color1.png -------------------------------------------------------------------------------- /lecture02/eyediagram.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture02/eyediagram.gif -------------------------------------------------------------------------------- /lecture02/hsv1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture02/hsv1.png -------------------------------------------------------------------------------- /lecture02/lecture02.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture02/lecture02.pdf -------------------------------------------------------------------------------- /lecture02/rgb1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture02/rgb1.png -------------------------------------------------------------------------------- /lecture02/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput 302 | -------------------------------------------------------------------------------- /lecture02/wb.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture02/wb.png -------------------------------------------------------------------------------- /lecture02/xyz1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture02/xyz1.png -------------------------------------------------------------------------------- /lecture03/bibliography.bib: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /lecture03/lecture03.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture03/lecture03.pdf -------------------------------------------------------------------------------- /lecture03/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput 302 | -------------------------------------------------------------------------------- /lecture04/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{krishna2017visual, 2 | title={Visual genome: Connecting language and vision using crowdsourced dense image annotations}, 3 | author={Krishna, Ranjay and Zhu, Yuke and Groth, Oliver and Johnson, Justin and Hata, Kenji and Kravitz, Joshua and Chen, Stephanie and Kalantidis, Yannis and Li, Li-Jia and Shamma, David A and others}, 4 | journal={International Journal of Computer Vision}, 5 | volume={123}, 6 | number={1}, 7 | pages={32--73}, 8 | year={2017}, 9 | publisher={Springer} 10 | } 11 | -------------------------------------------------------------------------------- /lecture04/histogram.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture04/histogram.png -------------------------------------------------------------------------------- /lecture04/impulse_response.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture04/impulse_response.jpeg -------------------------------------------------------------------------------- /lecture04/input_signal_.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture04/input_signal_.jpeg -------------------------------------------------------------------------------- /lecture04/lecture04.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture04/lecture04.pdf -------------------------------------------------------------------------------- /lecture04/res.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture04/res.png -------------------------------------------------------------------------------- /lecture04/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput 302 | -------------------------------------------------------------------------------- /lecture05/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{walther2011simple, 2 | title={Simple line drawings suffice for functional MRI decoding of natural scene categories}, 3 | author={Walther, Dirk B and Chai, Barry and Caddigan, Eamon and Beck, Diane M and Fei-Fei, Li}, 4 | journal={Proceedings of the National Academy of Sciences}, 5 | volume={108}, 6 | number={23}, 7 | pages={9661--9666}, 8 | year={2011}, 9 | publisher={National Acad Sciences} 10 | } 11 | @article{hubel1960receptive, 12 | title={Receptive fields of optic nerve fibres in the spider monkey}, 13 | author={Hubel, DH and Wiesel, TN}, 14 | journal={The Journal of physiology}, 15 | volume={154}, 16 | number={3}, 17 | pages={572--580}, 18 | year={1960}, 19 | publisher={Wiley Online Library} 20 | } 21 | -------------------------------------------------------------------------------- /lecture05/biederman_outlines.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/biederman_outlines.png -------------------------------------------------------------------------------- /lecture05/fei-fei_detection.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/fei-fei_detection.png -------------------------------------------------------------------------------- /lecture05/fei-fei_images.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/fei-fei_images.png -------------------------------------------------------------------------------- /lecture05/hubel_wiesel_cat.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/hubel_wiesel_cat.jpg -------------------------------------------------------------------------------- /lecture05/lazebnik.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/lazebnik.png -------------------------------------------------------------------------------- /lecture05/lazebnik_tiger.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/lazebnik_tiger.png -------------------------------------------------------------------------------- /lecture05/lecture05.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/lecture05.pdf -------------------------------------------------------------------------------- /lecture05/seitz_gradient_direction.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/seitz_gradient_direction.png -------------------------------------------------------------------------------- /lecture05/seitz_noise.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/seitz_noise.png -------------------------------------------------------------------------------- /lecture05/seitz_smoothing.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture05/seitz_smoothing.png -------------------------------------------------------------------------------- /lecture05/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput 302 | -------------------------------------------------------------------------------- /lecture06/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{line_drawings, 2 | title={Simple line drawings suffice for functional MRI 3 | decoding of natural scene categories}, 4 | author={Dirk B. Walther and Barry Chai and Eamon Caddigan and Diane M. Beck and Li Fei-Fei}, 5 | journal={Proceedings of the National Academy of Sciences}, 6 | volume={108}, 7 | number={23}, 8 | pages={9661–-9666}, 9 | year={2011}, 10 | publisher={National Academy of Sciences} 11 | } 12 | @article{fischler1981random, 13 | title={Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography}, 14 | author={Fischler, Martin A and Bolles, Robert C}, 15 | journal={Communications of the ACM}, 16 | volume={24}, 17 | number={6}, 18 | pages={381--395}, 19 | year={1981}, 20 | publisher={ACM} 21 | } 22 | @book{prince2012computer, 23 | title={Computer vision: models, learning, and inference}, 24 | author={Prince, Simon JD}, 25 | year={2012}, 26 | publisher={Cambridge University Press} 27 | } 28 | @book{forsyth2011computer, 29 | title={Computer vision: a modern approach}, 30 | author={Forsyth, David and Ponce, Jean}, 31 | year={2011}, 32 | publisher={Upper Saddle River, NJ; London: Prentice Hall} 33 | } 34 | -------------------------------------------------------------------------------- /lecture06/hough_transform.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/hough_transform.png -------------------------------------------------------------------------------- /lecture06/hough_transform2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/hough_transform2.png -------------------------------------------------------------------------------- /lecture06/hough_transform3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/hough_transform3.png -------------------------------------------------------------------------------- /lecture06/hough_transform4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/hough_transform4.png -------------------------------------------------------------------------------- /lecture06/lecture06.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/lecture06.pdf -------------------------------------------------------------------------------- /lecture06/line_drawings.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/line_drawings.png -------------------------------------------------------------------------------- /lecture06/overall.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/overall.png -------------------------------------------------------------------------------- /lecture06/ransac1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/ransac1.png -------------------------------------------------------------------------------- /lecture06/ransac2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/ransac2.png -------------------------------------------------------------------------------- /lecture06/ransac3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/ransac3.png -------------------------------------------------------------------------------- /lecture06/ransac4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture06/ransac4.png -------------------------------------------------------------------------------- /lecture06/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput 302 | -------------------------------------------------------------------------------- /lecture07/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{krishna2017visual, 2 | title={Visual genome: Connecting language and vision using crowdsourced dense image annotations}, 3 | author={Krishna, Ranjay and Zhu, Yuke and Groth, Oliver and Johnson, Justin and Hata, Kenji and Kravitz, Joshua and Chen, Stephanie and Kalantidis, Yannis and Li, Li-Jia and Shamma, David A and others}, 4 | journal={International Journal of Computer Vision}, 5 | volume={123}, 6 | number={1}, 7 | pages={32--73}, 8 | year={2017}, 9 | publisher={Springer} 10 | } 11 | 12 | @book{opencv, title={OpenCV 3.1.0 Documentation: $https://docs.opencv.org/3.1.0/$}, url={https://docs.opencv.org/3.1.0/sift_scale_invariant.jpg}, journal={OpenCV}} 13 | -------------------------------------------------------------------------------- /lecture07/eigenvalues_harris.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture07/eigenvalues_harris.png -------------------------------------------------------------------------------- /lecture07/harris_response.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture07/harris_response.png -------------------------------------------------------------------------------- /lecture07/lecture07.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture07/lecture07.pdf -------------------------------------------------------------------------------- /lecture07/local_feature_ex.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture07/local_feature_ex.png -------------------------------------------------------------------------------- /lecture07/scaleinvariant.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture07/scaleinvariant.png -------------------------------------------------------------------------------- /lecture07/sift_scale_invariant.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture07/sift_scale_invariant.jpg -------------------------------------------------------------------------------- /lecture08/bibliography.bib: -------------------------------------------------------------------------------- 1 | title={Distinctive image features from scale-invariant keypoints}, 2 | author={Lowe, David}, 3 | journal={International Journal of Computer Vision}, 4 | volume={60}, 5 | number={2}, 6 | pages={91--110}, 7 | year={2004} 8 | } 9 | 10 | title={Detection of local features invariant to 11 | affine transformations}, 12 | author={Mikolajczyk, Krystian}, 13 | journal={International Journal of Computer Vision}, 14 | volume={60}, 15 | number={2}, 16 | pages={91--110}, 17 | year={2002} 18 | } 19 | 20 | -------------------------------------------------------------------------------- /lecture08/bike.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/bike.png -------------------------------------------------------------------------------- /lecture08/cmu_composite_panorama.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/cmu_composite_panorama.png -------------------------------------------------------------------------------- /lecture08/corners.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/corners.png -------------------------------------------------------------------------------- /lecture08/dog_pyramid.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/dog_pyramid.jpg -------------------------------------------------------------------------------- /lecture08/histogram.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/histogram.png -------------------------------------------------------------------------------- /lecture08/histogramsensitivity.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/histogramsensitivity.png -------------------------------------------------------------------------------- /lecture08/homography_prob.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/homography_prob.png -------------------------------------------------------------------------------- /lecture08/lecture08.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/lecture08.pdf -------------------------------------------------------------------------------- /lecture08/max.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/max.png -------------------------------------------------------------------------------- /lecture08/maxima.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/maxima.png -------------------------------------------------------------------------------- /lecture08/transMatrices.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/transMatrices.png -------------------------------------------------------------------------------- /lecture08/transforms.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture08/transforms.png -------------------------------------------------------------------------------- /lecture09/Function.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/Function.png -------------------------------------------------------------------------------- /lecture09/Naive_Castle_Resizing.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/Naive_Castle_Resizing.jpg -------------------------------------------------------------------------------- /lecture09/Optimal_Seam.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/Optimal_Seam.png -------------------------------------------------------------------------------- /lecture09/Pixel_Removal.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/Pixel_Removal.png -------------------------------------------------------------------------------- /lecture09/Smart_Resizing.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/Smart_Resizing.jpg -------------------------------------------------------------------------------- /lecture09/Tracking.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/Tracking.png -------------------------------------------------------------------------------- /lecture09/backtrack.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/backtrack.png -------------------------------------------------------------------------------- /lecture09/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{siggraphseamcarving, 2 | title={Seam carving for content-aware image resizing}, 3 | author={Shai Avidan, Ariel Shamir}, 4 | journal={ACM Trans. Graph}, 5 | volume={26}, 6 | number={3}, 7 | pages={10}, 8 | year={2007}, 9 | } 10 | @inproceedings{judd2009learning, 11 | title={Learning to predict where humans look}, 12 | author={Judd, Tilke and Ehinger, Krista and Durand, Fr{\'e}do and Torralba, Antonio}, 13 | booktitle={Computer Vision, 2009 IEEE 12th international conference on}, 14 | pages={2106--2113}, 15 | year={2009}, 16 | organization={IEEE} 17 | } 18 | @inproceedings{castle, 19 | title={https://commons.wikimedia.org/wiki/File:Broadway\_tower\_edit.jpg}, 20 | author={}, 21 | booktitle={}, 22 | pages={}, 23 | year={}, 24 | organization={} 25 | } 26 | @inproceedings{castle2, 27 | title={https://commons.wikimedia.org/wiki/File:Broadway\_tower\_edit.jpg}, 28 | author={}, 29 | booktitle={}, 30 | pages={}, 31 | year={}, 32 | organization={} 33 | } 34 | @inproceedings{avidan2007seam, 35 | title={Seam carving for content-aware image resizing}, 36 | author={Avidan, Shai and Shamir, Ariel}, 37 | booktitle={ACM Transactions on graphics (TOG)}, 38 | volume={26}, 39 | number={3}, 40 | pages={10}, 41 | year={2007}, 42 | organization={ACM} 43 | } 44 | 45 | @inproceedings{rubinstein2008improved, 46 | title={Improved seam carving for video retargeting}, 47 | author={Rubinstein, Michael and Shamir, Ariel and Avidan, Shai}, 48 | booktitle={ACM transactions on graphics (TOG)}, 49 | volume={27}, 50 | number={3}, 51 | pages={16}, 52 | year={2008}, 53 | organization={ACM} 54 | } 55 | @inproceedings{illusion, 56 | title={https://www.weirdoptics.com/hidden-lion-visual-optical-illusion}, 57 | author={}, 58 | booktitle={}, 59 | pages={}, 60 | year={}, 61 | organization={} 62 | } 63 | 64 | @inproceedings{ren2003learning, 65 | title={Learning a classification model for segmentation}, 66 | author={Ren, Xiaofeng and Malik, Jitendra}, 67 | booktitle={null}, 68 | pages={10}, 69 | year={2003}, 70 | organization={IEEE} 71 | } 72 | -------------------------------------------------------------------------------- /lecture09/camels.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/camels.png -------------------------------------------------------------------------------- /lecture09/content_aware_resizing.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/content_aware_resizing.png -------------------------------------------------------------------------------- /lecture09/energy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/energy.png -------------------------------------------------------------------------------- /lecture09/findseam1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/findseam1.png -------------------------------------------------------------------------------- /lecture09/forward_energy.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/forward_energy.PNG -------------------------------------------------------------------------------- /lecture09/forward_energy_calculation.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/forward_energy_calculation.JPG -------------------------------------------------------------------------------- /lecture09/lecture09.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/lecture09.pdf -------------------------------------------------------------------------------- /lecture09/limitations_close-up.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/limitations_close-up.JPG -------------------------------------------------------------------------------- /lecture09/lion.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/lion.jpg -------------------------------------------------------------------------------- /lecture09/main.synctex(busy): -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/main.synctex(busy) -------------------------------------------------------------------------------- /lecture09/muller.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/muller.jpg -------------------------------------------------------------------------------- /lecture09/object_removal.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/object_removal.JPG -------------------------------------------------------------------------------- /lecture09/objects.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/objects.png -------------------------------------------------------------------------------- /lecture09/people.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/people.png -------------------------------------------------------------------------------- /lecture09/resizing_representation.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/resizing_representation.JPG -------------------------------------------------------------------------------- /lecture09/seam_carving_limitations.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/seam_carving_limitations.JPG -------------------------------------------------------------------------------- /lecture09/superpixels.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/superpixels.png -------------------------------------------------------------------------------- /lecture09/user_constraints.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/user_constraints.PNG -------------------------------------------------------------------------------- /lecture09/video_retargeting.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture09/video_retargeting.JPG -------------------------------------------------------------------------------- /lecture10/average.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/average.png -------------------------------------------------------------------------------- /lecture10/bibliography.bib: -------------------------------------------------------------------------------- 1 | @book{forsyth2011computer, 2 | title={Computer vision: a modern approach}, 3 | author={Forsyth, David and Ponce, Jean}, 4 | year={2011}, 5 | publisher={Upper Saddle River, NJ; London: Prentice Hall} 6 | } 7 | -------------------------------------------------------------------------------- /lecture10/clustering_for_summarization.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/clustering_for_summarization.png -------------------------------------------------------------------------------- /lecture10/complete.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/complete.png -------------------------------------------------------------------------------- /lecture10/continuity-occlusion-2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/continuity-occlusion-2.png -------------------------------------------------------------------------------- /lecture10/continuity-occlusion.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/continuity-occlusion.png -------------------------------------------------------------------------------- /lecture10/dendrogram2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/dendrogram2.png -------------------------------------------------------------------------------- /lecture10/figure-ground.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/figure-ground.png -------------------------------------------------------------------------------- /lecture10/gestalt-factors.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/gestalt-factors.png -------------------------------------------------------------------------------- /lecture10/k-means-algorithm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/k-means-algorithm.png -------------------------------------------------------------------------------- /lecture10/k-means-example.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/k-means-example.png -------------------------------------------------------------------------------- /lecture10/lecture10.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/lecture10.pdf -------------------------------------------------------------------------------- /lecture10/mean-shift.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/mean-shift.png -------------------------------------------------------------------------------- /lecture10/panda.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/panda.jpg -------------------------------------------------------------------------------- /lecture10/single.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/single.png -------------------------------------------------------------------------------- /lecture10/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput -------------------------------------------------------------------------------- /lecture10/tiger-segmented.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/tiger-segmented.png -------------------------------------------------------------------------------- /lecture10/tiger.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture10/tiger.png -------------------------------------------------------------------------------- /lecture11/Bias-VarianceTrade-Off.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture11/Bias-VarianceTrade-Off.png -------------------------------------------------------------------------------- /lecture11/ChoosingValueOfK.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture11/ChoosingValueOfK.png -------------------------------------------------------------------------------- /lecture11/DimensionalityIssue.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture11/DimensionalityIssue.png -------------------------------------------------------------------------------- /lecture11/EuclideanIssue.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture11/EuclideanIssue.png -------------------------------------------------------------------------------- /lecture11/basin1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture11/basin1.png -------------------------------------------------------------------------------- /lecture11/basin2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture11/basin2.png -------------------------------------------------------------------------------- /lecture11/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{krishna2017visual, 2 | title={Visual genome: Connecting language and vision using crowdsourced dense image annotations}, 3 | author={Krishna, Ranjay and Zhu, Yuke and Groth, Oliver and Johnson, Justin and Hata, Kenji and Kravitz, Joshua and Chen, Stephanie and Kalantidis, Yannis and Li, Li-Jia and Shamma, David A and others}, 4 | journal={International Journal of Computer Vision}, 5 | volume={123}, 6 | number={1}, 7 | pages={32--73}, 8 | year={2017}, 9 | publisher={Springer} 10 | } 11 | 12 | @electronic{lecture11, 13 | author={Niebles, Juan Carlos and Krishna, Ranjay}, 14 | institution= {Stanford University}, 15 | month = {October}, 16 | year = {2017}, 17 | title= {Lecture 11: Object recognition} 18 | } 19 | -------------------------------------------------------------------------------- /lecture11/decision_boudnaries.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture11/decision_boudnaries.jpg -------------------------------------------------------------------------------- /lecture11/knearestneighbors.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture11/knearestneighbors.png -------------------------------------------------------------------------------- /lecture11/lecture11.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture11/lecture11.pdf -------------------------------------------------------------------------------- /lecture12/Dimension_reduction_High_Level.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/Dimension_reduction_High_Level.png -------------------------------------------------------------------------------- /lecture12/PCA_dim.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/PCA_dim.png -------------------------------------------------------------------------------- /lecture12/Screen_Shot_2017-11-08_at_8_29_32_PM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/Screen_Shot_2017-11-08_at_8_29_32_PM.png -------------------------------------------------------------------------------- /lecture12/Usig.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/Usig.PNG -------------------------------------------------------------------------------- /lecture12/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{moore-penrose, 2 | title={The Moore-Penrose Inverse and Least Squares}, 3 | author={MacAusland, R.}, 4 | year={2014}, 5 | organization={University of Puget Sound} 6 | } 7 | 8 | @article{search-clustering, 9 | title={Search Result Clustering using a Singular Value Decomposition (SVD)}, 10 | author={Abdulla H.D., Snasel V.}, 11 | year={2009}, 12 | publisher={Springer}, 13 | organization={First International Conference on Intelligent Human Computer Interaction} 14 | } 15 | 16 | @inproceedings{ams, 17 | title={http://www.ams.org/samplings/feature-column/fcarc-svd}, 18 | author={}, 19 | booktitle={}, 20 | pages={}, 21 | year={}, 22 | organization={} 23 | } 24 | -------------------------------------------------------------------------------- /lecture12/compare.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/compare.png -------------------------------------------------------------------------------- /lecture12/full_pca.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/full_pca.PNG -------------------------------------------------------------------------------- /lecture12/lecture12.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/lecture12.pdf -------------------------------------------------------------------------------- /lecture12/svd_first10.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/svd_first10.png -------------------------------------------------------------------------------- /lecture12/svd_image_compress.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/svd_image_compress.png -------------------------------------------------------------------------------- /lecture12/svd_original.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/svd_original.png -------------------------------------------------------------------------------- /lecture12/usigv_example.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture12/usigv_example.PNG -------------------------------------------------------------------------------- /lecture13/between-within-scatters.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/between-within-scatters.png -------------------------------------------------------------------------------- /lecture13/disgust.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/disgust.png -------------------------------------------------------------------------------- /lecture13/eigenvalues.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/eigenvalues.png -------------------------------------------------------------------------------- /lecture13/error.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/error.png -------------------------------------------------------------------------------- /lecture13/error_rate.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/error_rate.png -------------------------------------------------------------------------------- /lecture13/face_non-face_plot.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/face_non-face_plot.png -------------------------------------------------------------------------------- /lecture13/generic.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/generic.png -------------------------------------------------------------------------------- /lecture13/happy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/happy.png -------------------------------------------------------------------------------- /lecture13/lecture13.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/lecture13.pdf -------------------------------------------------------------------------------- /lecture13/lecture13.synctex.gz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/lecture13.synctex.gz -------------------------------------------------------------------------------- /lecture13/pca-vs-lda.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/pca-vs-lda.png -------------------------------------------------------------------------------- /lecture13/reconstructed.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/reconstructed.png -------------------------------------------------------------------------------- /lecture13/sample.bib: -------------------------------------------------------------------------------- 1 | @article{eigenfaces, 2 | author = { Matthew Turk and Alex Pentland}, 3 | title = {Eigenfaces for Recognition}, 4 | journal = {Journal of Cognitive Neuroscience}, 5 | volume = {3}, 6 | number = {1}, 7 | pages = {71-86}, 8 | year = {1991} 9 | } 10 | @article{EigenfacesvsFisherfaces , 11 | title={Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection}, 12 | author={P. Belhumeur, J. Hespanha, and D. Kriegman}, 13 | journal={IEEE Transactions on pattern analysis and machine intelligence}, 14 | volume={19}, 15 | number={7}, 16 | pages={711-720}, 17 | year={1997}, 18 | } -------------------------------------------------------------------------------- /lecture13/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput -------------------------------------------------------------------------------- /lecture13/variation_in_expression.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture13/variation_in_expression.png -------------------------------------------------------------------------------- /lecture14/Bayes_Histogram.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/Bayes_Histogram.png -------------------------------------------------------------------------------- /lecture14/BoW_pyramid.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/BoW_pyramid.PNG -------------------------------------------------------------------------------- /lecture14/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{Large-scale-collections, 2 | title={Learning to Match Images in Large-Scale Collections}, 3 | author={Song Cao, Noah Snavely}, 4 | publisher={}, 5 | organization={Cornell University} 6 | } 7 | @article{vocab-tree, 8 | title={Scalable Recognition with a Vocabulary Tree}, 9 | author={David Nistér, Henrik Stewénius}, 10 | publisher={}, 11 | organization={Center for Visualization and Virtual Environments, University of Kentucky} 12 | } 13 | 14 | @misc{slides, 15 | title={Lecture: Visual Bag of Words}, 16 | author={Juan Carlos Niebles, Ranjay Krishna}, 17 | howpublished = "\url{http://vision.stanford.edu/teaching/cs131_fall1718/files/14_BoW_bayes.pdf}", 18 | publisher={}, 19 | organization={Stanford University} 20 | } 21 | 22 | @misc{pyramid, 23 | author = {Wikipedia}, 24 | title = {{Pyramid (image processing)}}, 25 | howpublished = "\url{https://en.wikipedia.org/wiki/Pyramid_(image_processing)}", 26 | note = "[Online; accessed 15-Nov-2017]" 27 | } 28 | 29 | @misc{quora, 30 | author = {Tomas Mardones}, 31 | title = {{How should we understand the Spatial Pyramid Matching?}}, 32 | howpublished = "\url{https://www.quora.com/How-should-we-understand-the-Spatial-Pyramid-Matching}", 33 | note = "[Online; accessed 15-Nov-2017]" 34 | } 35 | 36 | 37 | -------------------------------------------------------------------------------- /lecture14/bow.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/bow.png -------------------------------------------------------------------------------- /lecture14/grid_features.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/grid_features.png -------------------------------------------------------------------------------- /lecture14/hist.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/hist.png -------------------------------------------------------------------------------- /lecture14/lecture14.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/lecture14.pdf -------------------------------------------------------------------------------- /lecture14/pyramid.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/pyramid.PNG -------------------------------------------------------------------------------- /lecture14/pyramid_data1.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/pyramid_data1.PNG -------------------------------------------------------------------------------- /lecture14/pyramid_data2.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/pyramid_data2.PNG -------------------------------------------------------------------------------- /lecture14/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput -------------------------------------------------------------------------------- /lecture14/visual_vocab.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture14/visual_vocab.png -------------------------------------------------------------------------------- /lecture15/average_face_template.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/average_face_template.png -------------------------------------------------------------------------------- /lecture15/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{hog_human_detection, 2 | title={Histograms of Oriented Gradients for Human Detection}, 3 | author={Dalal Navneet, Bill Triggs}, 4 | journal={Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on}, 5 | publisher={IEEE}, 6 | year={2005}, 7 | volume={1} 8 | } 9 | 10 | @article{DPM, 11 | title={The Representation and Matching of Pictorial Structures}, 12 | author={Martin A. Fischler, Robert A. Elshlager}, 13 | journal={IEEE Transactions on Computers ( Volume: C-22, Issue: 1, Jan. 1973 )}, 14 | publisher={IEEE}, 15 | year={1973}, 16 | } 17 | 18 | @misc{pascal-voc-2012, 19 | author = "Everingham, M. and Van~Gool, L. and Williams, C. K. I. and Winn, J. and Zisserman, A.", 20 | title = "The {PASCAL} {V}isual {O}bject {C}lasses {C}hallenge 2012 {(VOC2012)} {R}esults", 21 | howpublished = "http://www.pascal-network.org/challenges/VOC/voc2012/workshop/index.html" 22 | } 23 | 24 | @article{ILSVRC15, 25 | Author = {Olga Russakovsky and Jia Deng and Hao Su and Jonathan Krause and Sanjeev Satheesh and Sean Ma and Zhiheng Huang and Andrej Karpathy and Aditya Khosla and Michael Bernstein and Alexander C. Berg and Li Fei-Fei}, 26 | Title = {{ImageNet Large Scale Visual Recognition Challenge}}, 27 | Year = {2015}, 28 | journal = {International Journal of Computer Vision (IJCV)}, 29 | doi = {10.1007/s11263-015-0816-y}, 30 | volume={115}, 31 | number={3}, 32 | pages={211-252} 33 | } 34 | 35 | @article{DBLP:journals/corr/LinMBHPRDZ14, 36 | author = {Tsung{-}Yi Lin and 37 | Michael Maire and 38 | Serge J. Belongie and 39 | Lubomir D. Bourdev and 40 | Ross B. Girshick and 41 | James Hays and 42 | Pietro Perona and 43 | Deva Ramanan and 44 | Piotr Doll{\'{a}}r and 45 | C. Lawrence Zitnick}, 46 | title = {Microsoft {COCO:} Common Objects in Context}, 47 | journal = {CoRR}, 48 | volume = {abs/1405.0312}, 49 | year = {2014}, 50 | url = {http://arxiv.org/abs/1405.0312}, 51 | archivePrefix = {arXiv}, 52 | eprint = {1405.0312}, 53 | timestamp = {Wed, 07 Jun 2017 14:41:35 +0200}, 54 | biburl = {http://dblp.org/rec/bib/journals/corr/LinMBHPRDZ14}, 55 | bibsource = {dblp computer science bibliography, http://dblp.org} 56 | } 57 | -------------------------------------------------------------------------------- /lecture15/car_model.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/car_model.png -------------------------------------------------------------------------------- /lecture15/chart.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/chart.jpg -------------------------------------------------------------------------------- /lecture15/coco.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/coco.png -------------------------------------------------------------------------------- /lecture15/deformable_bike_model.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/deformable_bike_model.png -------------------------------------------------------------------------------- /lecture15/deformable_head_filter.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/deformable_head_filter.png -------------------------------------------------------------------------------- /lecture15/deformable_model.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/deformable_model.png -------------------------------------------------------------------------------- /lecture15/deformable_score1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/deformable_score1.png -------------------------------------------------------------------------------- /lecture15/false_neg.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/false_neg.jpg -------------------------------------------------------------------------------- /lecture15/false_pos.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/false_pos.jpg -------------------------------------------------------------------------------- /lecture15/feature_pyramid.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/feature_pyramid.jpg -------------------------------------------------------------------------------- /lecture15/fullyconnected.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/fullyconnected.png -------------------------------------------------------------------------------- /lecture15/ilsvr.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/ilsvr.png -------------------------------------------------------------------------------- /lecture15/lecture15.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/lecture15.pdf -------------------------------------------------------------------------------- /lecture15/object-detector-intro.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/object-detector-intro.png -------------------------------------------------------------------------------- /lecture15/person_star.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/person_star.png -------------------------------------------------------------------------------- /lecture15/person_template.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/person_template.png -------------------------------------------------------------------------------- /lecture15/pipeline.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/pipeline.png -------------------------------------------------------------------------------- /lecture15/pipeline2a.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/pipeline2a.png -------------------------------------------------------------------------------- /lecture15/pipeline2b.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/pipeline2b.png -------------------------------------------------------------------------------- /lecture15/pipelinescoreplot.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/pipelinescoreplot.png -------------------------------------------------------------------------------- /lecture15/prec_recall_pic.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/prec_recall_pic.jpg -------------------------------------------------------------------------------- /lecture15/precision_recall.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/precision_recall.jpg -------------------------------------------------------------------------------- /lecture15/predict_truth.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/predict_truth.jpg -------------------------------------------------------------------------------- /lecture15/results-tpfp.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/results-tpfp.png -------------------------------------------------------------------------------- /lecture15/sliding_window_a.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/sliding_window_a.jpg -------------------------------------------------------------------------------- /lecture15/sliding_window_b.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/sliding_window_b.jpg -------------------------------------------------------------------------------- /lecture15/sliding_window_c.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/sliding_window_c.jpg -------------------------------------------------------------------------------- /lecture15/sliding_window_d.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/sliding_window_d.jpg -------------------------------------------------------------------------------- /lecture15/small_sliding_window_a.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/small_sliding_window_a.jpg -------------------------------------------------------------------------------- /lecture15/small_sliding_window_b.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/small_sliding_window_b.jpg -------------------------------------------------------------------------------- /lecture15/star.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/star.png -------------------------------------------------------------------------------- /lecture15/true_pos.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture15/true_pos.jpg -------------------------------------------------------------------------------- /lecture16/bo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture16/bo.png -------------------------------------------------------------------------------- /lecture16/bubblestudy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture16/bubblestudy.png -------------------------------------------------------------------------------- /lecture16/confusionmatrix.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture16/confusionmatrix.png -------------------------------------------------------------------------------- /lecture16/fbyoutubedata.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture16/fbyoutubedata.png -------------------------------------------------------------------------------- /lecture16/heatmap.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture16/heatmap.png -------------------------------------------------------------------------------- /lecture16/internettraffic.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture16/internettraffic.png -------------------------------------------------------------------------------- /lecture16/lecture16.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture16/lecture16.pdf -------------------------------------------------------------------------------- /lecture16/obamacreditcard.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture16/obamacreditcard.png -------------------------------------------------------------------------------- /lecture16/pascal.bib: -------------------------------------------------------------------------------- 1 | @misc{pascal-voc-2012, 2 | author = "Everingham, M. and Van~Gool, L. and Williams, C. K. I. and Winn, J. and Zisserman, A.", 3 | title = "The {PASCAL} {V}isual {O}bject {C}lasses {C}hallenge 2012 {(VOC2012)} {R}esults", 4 | howpublished = "http://www.pascal-network.org/challenges/VOC/voc2012/workshop/index.html"} -------------------------------------------------------------------------------- /lecture16/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput -------------------------------------------------------------------------------- /lecture17/aperture-problem.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/aperture-problem.jpg -------------------------------------------------------------------------------- /lecture17/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{krishna2017visual, 2 | title={Visual genome: Connecting language and vision using crowdsourced dense image annotations}, 3 | author={Krishna, Ranjay and Zhu, Yuke and Groth, Oliver and Johnson, Justin and Hata, Kenji and Kravitz, Joshua and Chen, Stephanie and Kalantidis, Yannis and Li, Li-Jia and Shamma, David A and others}, 4 | journal={International Journal of Computer Vision}, 5 | volume={123}, 6 | number={1}, 7 | pages={32--73}, 8 | year={2017}, 9 | publisher={Springer} 10 | } 11 | 12 | @incollection{fleet2006optical, 13 | title={Optical flow estimation}, 14 | author={Fleet, David and Weiss, Yair}, 15 | booktitle={Handbook of mathematical models in computer vision}, 16 | pages={237--257}, 17 | year={2006}, 18 | publisher={Springer} 19 | } 20 | 21 | @article{lucas1981iterative, 22 | title={An iterative image registration technique with an application to stereo vision}, 23 | author={Lucas, Bruce D and Kanade, Takeo and others}, 24 | year={1981}, 25 | publisher={Vancouver, BC, Canada} 26 | } 27 | -------------------------------------------------------------------------------- /lecture17/grouping.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/grouping.png -------------------------------------------------------------------------------- /lecture17/lecture17.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/lecture17.pdf -------------------------------------------------------------------------------- /lecture17/lk-pyr.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/lk-pyr.png -------------------------------------------------------------------------------- /lecture17/lk_eigen.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/lk_eigen.png -------------------------------------------------------------------------------- /lecture17/lk_regions.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/lk_regions.png -------------------------------------------------------------------------------- /lecture17/lucas-kanade-meme.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/lucas-kanade-meme.jpg -------------------------------------------------------------------------------- /lecture17/michaelblack.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/michaelblack.png -------------------------------------------------------------------------------- /lecture17/quadratic.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/quadratic.png -------------------------------------------------------------------------------- /lecture17/treeLayer.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/treeLayer.png -------------------------------------------------------------------------------- /lecture17/treeOriginal.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/treeOriginal.png -------------------------------------------------------------------------------- /lecture17/w_pyr.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/w_pyr.png -------------------------------------------------------------------------------- /lecture17/wo_pyr.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture17/wo_pyr.png -------------------------------------------------------------------------------- /lecture18/bibliography.bib: -------------------------------------------------------------------------------- 1 | @INPROCEEDINGS{7456615, 2 | author={S. Chandra and G. Sharma and S. Malhotra and D. Jha and A. P. Mittal}, 3 | title={Eye tracking based human computer interaction: Applications and their uses}, 4 | year={2015}, 5 | pages={1-5}, 6 | ISSN={}, 7 | month={Dec},}, 8 | 9 | @ARTICLE{1406476, 10 | author={A. Hampapur and L. Brown and J. Connell and A. Ekin and N. Haas and M. Lu and H. Merkl and S. Pankanti}, 11 | journal={IEEE Signal Processing Magazine}, 12 | title={Smart video surveillance: exploring the concept of multiscale spatiotemporal tracking}, 13 | year={2005}, 14 | volume={22}, 15 | number={2}, 16 | pages={38-51}, 17 | doi={10.1109/MSP.2005.1406476}, 18 | ISSN={1053-5888}, 19 | month={March},} 20 | , 21 | @ARTICLE{1634325, 22 | author={A. I. Comport and E. Marchand and M. Pressigout and F. Chaumette}, 23 | journal={IEEE Transactions on Visualization and Computer Graphics}, 24 | title={Real-time markerless tracking for augmented reality: the virtual visual servoing framework}, 25 | year={2006}, 26 | volume={12}, 27 | number={4}, 28 | pages={615-628}, 29 | doi={10.1109/TVCG.2006.78}, 30 | ISSN={1077-2626}, 31 | month={July},}, 32 | @INPROCEEDINGS{1545408, 33 | author={M. Singh and M. Mandal and A. Basu}, 34 | booktitle={2005 IEEE/RSJ International Conference on Intelligent Robots and Systems}, 35 | title={Visual gesture recognition for ground air traffic control using the Radon transform}, 36 | year={2005}, 37 | volume={}, 38 | number={}, 39 | pages={2586-2591}, 40 | doi={10.1109/IROS.2005.1545408}, 41 | ISSN={2153-0858}, 42 | month={Aug},} 43 | , 44 | @INPROCEEDINGS{4054748, 45 | author={O. Rostamianfar and F. Janabi-Sharifi and I. Hassanzadeh}, 46 | booktitle={2006 Canadian Conference on Electrical and Computer Engineering}, 47 | title={Visual Tracking System for Dense Traffic Intersections}, 48 | year={2006}, 49 | volume={}, 50 | number={}, 51 | pages={2000-2004}, 52 | doi={10.1109/CCECE.2006.277838}, 53 | ISSN={0840-7789}, 54 | month={May},}, 55 | @ARTICLE{5484176, 56 | author={P. Mountney and D. Stoyanov and G. Z. Yang}, 57 | journal={IEEE Signal Processing Magazine}, 58 | title={Three-Dimensional Tissue Deformation Recovery and Tracking}, 59 | year={2010}, 60 | volume={27}, 61 | number={4}, 62 | pages={14-24}, 63 | ISSN={1053-5888}, 64 | month={July},} 65 | -------------------------------------------------------------------------------- /lecture18/feature_track_example.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture18/feature_track_example.PNG -------------------------------------------------------------------------------- /lecture18/fish.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture18/fish.png -------------------------------------------------------------------------------- /lecture18/lecture18.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture18/lecture18.pdf -------------------------------------------------------------------------------- /lecture18/man_track.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture18/man_track.png -------------------------------------------------------------------------------- /lecture18/template.tex: -------------------------------------------------------------------------------- 1 | \documentclass{article} 2 | 3 | \usepackage[final]{style} 4 | \usepackage[utf8]{inputenc} % allow utf-8 input 5 | \usepackage[T1]{fontenc} % use 8-bit T1 fonts 6 | \usepackage{hyperref} % hyperlinks 7 | \usepackage{url} % simple URL typesetting 8 | \usepackage{booktabs} % professional-quality tables 9 | \usepackage{amsfonts} % blackboard math symbols 10 | \usepackage{nicefrac} % compact symbols for 1/2, etc. 11 | \usepackage{microtype} % microtypography 12 | \usepackage{verbatim} 13 | \usepackage{graphicx} % for figures 14 | 15 | \title{Lecture \#X: NAME OF LECTURE} 16 | 17 | \author{ 18 | Student1 name, student2 name, etc. \\ 19 | Department of Computer Science\\ 20 | Stanford University\\ 21 | Stanford, CA 94305 \\ 22 | \texttt{\{STUDENT1, STUDENT2, etc.\}@cs.stanford.edu} \\ 23 | } 24 | 25 | \begin{document} 26 | 27 | \maketitle 28 | 29 | 30 | \section{Introduction} 31 | Copy this template and use it to write up the lecture notes. Also copy over 32 | bibliography.bib and add any references you use to that file. 33 | 34 | % References 35 | \small 36 | \bibliographystyle{plain} 37 | \bibliography{bibliography} 38 | \end{document} 39 | -------------------------------------------------------------------------------- /lecture18/trans_types.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture18/trans_types.png -------------------------------------------------------------------------------- /lecture18/translation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture18/translation.png -------------------------------------------------------------------------------- /lecture19/Activation_map.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/Activation_map.png -------------------------------------------------------------------------------- /lecture19/CNN_stacking.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/CNN_stacking.png -------------------------------------------------------------------------------- /lecture19/Conventional_schematic.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/Conventional_schematic.png -------------------------------------------------------------------------------- /lecture19/DL_schematic.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/DL_schematic.png -------------------------------------------------------------------------------- /lecture19/Hierarchical_features.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/Hierarchical_features.png -------------------------------------------------------------------------------- /lecture19/Inception.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/Inception.png -------------------------------------------------------------------------------- /lecture19/Linear_regression.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/Linear_regression.png -------------------------------------------------------------------------------- /lecture19/activation_functions.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/activation_functions.png -------------------------------------------------------------------------------- /lecture19/activation_functions2.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/activation_functions2.PNG -------------------------------------------------------------------------------- /lecture19/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{krishna2017visual, 2 | title={Visual genome: Connecting language and vision using crowdsourced dense image annotations}, 3 | author={Krishna, Ranjay and Zhu, Yuke and Groth, Oliver and Johnson, Justin and Hata, Kenji and Kravitz, Joshua and Chen, Stephanie and Kalantidis, Yannis and Li, Li-Jia and Shamma, David A and others}, 4 | journal={International Journal of Computer Vision}, 5 | volume={123}, 6 | number={1}, 7 | pages={32--73}, 8 | year={2017}, 9 | publisher={Springer} 10 | } 11 | @article{DBLP:journals/corr/SzegedyLJSRAEVR14, 12 | author = {Christian Szegedy and 13 | Wei Liu and 14 | Yangqing Jia and 15 | Pierre Sermanet and 16 | Scott E. Reed and 17 | Dragomir Anguelov and 18 | Dumitru Erhan and 19 | Vincent Vanhoucke and 20 | Andrew Rabinovich}, 21 | title = {Going Deeper with Convolutions}, 22 | journal = {CoRR}, 23 | volume = {abs/1409.4842}, 24 | year = {2014}, 25 | url = {http://arxiv.org/abs/1409.4842}, 26 | archivePrefix = {arXiv}, 27 | eprint = {1409.4842}, 28 | timestamp = {Wed, 07 Jun 2017 14:40:42 +0200}, 29 | biburl = {http://dblp.org/rec/bib/journals/corr/SzegedyLJSRAEVR14}, 30 | bibsource = {dblp computer science bibliography, http://dblp.org} 31 | } 32 | 33 | @article{hof_2016, title={Deep Learning}, howpublished={https://www.technologyreview.com/s/513696/deep-learning/}, journal={MIT Technology Review}, publisher={MIT Technology Review}, author={Hof, Robert D.}, year={2016}, month={Mar}} 34 | 35 | @inproceedings{deng2009imagenet, 36 | title={Imagenet: A large-scale hierarchical image database}, 37 | author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li}, 38 | booktitle={Computer Vision and Pattern Recognition, 2009. CVPR 2009. IEEE Conference on}, 39 | pages={248--255}, 40 | year={2009}, 41 | organization={IEEE} 42 | } 43 | 44 | @inproceedings{lin2011large, 45 | title={Large-scale image classification: fast feature extraction and svm training}, 46 | author={Lin, Yuanqing and Lv, Fengjun and Zhu, Shenghuo and Yang, Ming and Cour, Timothee and Yu, Kai and Cao, Liangliang and Huang, Thomas}, 47 | booktitle={Computer Vision and Pattern Recognition (CVPR), 2011 IEEE Conference on}, 48 | pages={1689--1696}, 49 | year={2011}, 50 | organization={IEEE} 51 | } 52 | 53 | @inproceedings{krizhevsky2012imagenet, 54 | title={Imagenet classification with deep convolutional neural networks}, 55 | author={Krizhevsky, Alex and Sutskever, Ilya and Hinton, Geoffrey E}, 56 | booktitle={Advances in neural information processing systems}, 57 | pages={1097--1105}, 58 | year={2012} 59 | } 60 | 61 | @article{simonyan2014very, 62 | title={Very deep convolutional networks for large-scale image recognition}, 63 | author={Simonyan, Karen and Zisserman, Andrew}, 64 | journal={arXiv preprint arXiv:1409.1556}, 65 | year={2014} 66 | } 67 | 68 | @inproceedings{szegedy2015going, 69 | title={Going deeper with convolutions}, 70 | author={Szegedy, Christian and Liu, Wei and Jia, Yangqing and Sermanet, Pierre and Reed, Scott and Anguelov, Dragomir and Erhan, Dumitru and Vanhoucke, Vincent and Rabinovich, Andrew}, 71 | booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition}, 72 | pages={1--9}, 73 | year={2015} 74 | } 75 | 76 | @inproceedings{he2016deep, 77 | title={Deep residual learning for image recognition}, 78 | author={He, Kaiming and Zhang, Xiangyu and Ren, Shaoqing and Sun, Jian}, 79 | booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition}, 80 | pages={770--778}, 81 | year={2016} 82 | } 83 | 84 | @article{russakovsky2015imagenet, 85 | title={Imagenet large scale visual recognition challenge}, 86 | author={Russakovsky, Olga and Deng, Jia and Su, Hao and Krause, Jonathan and Satheesh, Sanjeev and Ma, Sean and Huang, Zhiheng and Karpathy, Andrej and Khosla, Aditya and Bernstein, Michael and others}, 87 | journal={International Journal of Computer Vision}, 88 | volume={115}, 89 | number={3}, 90 | pages={211--252}, 91 | year={2015}, 92 | publisher={Springer} 93 | } 94 | 95 | @article{everingham2015pascal, 96 | title={The pascal visual object classes challenge: A retrospective}, 97 | author={Everingham, Mark and Eslami, SM Ali and Van Gool, Luc and Williams, Christopher KI and Winn, John and Zisserman, Andrew}, 98 | journal={International journal of computer vision}, 99 | volume={111}, 100 | number={1}, 101 | pages={98--136}, 102 | year={2015}, 103 | publisher={Springer} 104 | } 105 | 106 | @inproceedings{he2016deep, 107 | title={Deep residual learning for image recognition}, 108 | author={He, Kaiming and Zhang, Xiangyu and Ren, Shaoqing and Sun, Jian}, 109 | booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition}, 110 | pages={770--778}, 111 | year={2016} 112 | } 113 | 114 | @inproceedings{semseg, 115 | title={Instance-aware semantic segmentation via multi-task network cascades}, 116 | author={Dai, Jifeng and He, Kaiming and Sun, Jian}, 117 | booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition}, 118 | pages={3150--3158}, 119 | year={2016} 120 | } 121 | 122 | @article{poseest, 123 | title={Realtime multi-person 2d pose estimation using part affinity fields}, 124 | author={Cao, Zhe and Simon, Tomas and Wei, Shih-En and Sheikh, Yaser}, 125 | journal={arXiv preprint arXiv:1611.08050}, 126 | year={2016} 127 | } 128 | 129 | @inproceedings{singlecap, 130 | title={Deep visual-semantic alignments for generating image descriptions}, 131 | author={Karpathy, Andrej and Fei-Fei, Li}, 132 | booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition}, 133 | pages={3128--3137}, 134 | year={2015} 135 | } 136 | 137 | @inproceedings(densecap, 138 | title={Densecap: Fully convolutional localization networks for dense captioning}, 139 | author={Johnson, Justin and Karpathy, Andrej and Fei-Fei, Li}, 140 | booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition}, 141 | pages={4565--4574}, 142 | year={2016} 143 | } 144 | 145 | @article{superres, 146 | title={Photo-realistic single image super-resolution using a generative adversarial network}, 147 | author={Ledig, Christian and Theis, Lucas and Husz{\'a}r, Ferenc and Caballero, Jose and Cunningham, Andrew and Acosta, Alejandro and Aitken, Andrew and Tejani, Alykhan and Totz, Johannes and Wang, Zehan and others}, 148 | journal={arXiv preprint arXiv:1609.04802}, 149 | year={2016} 150 | } 151 | 152 | @inproceedings{vqa, 153 | title={Visual7w: Grounded question answering in images}, 154 | author={Zhu, Yuke and Groth, Oliver and Bernstein, Michael and Fei-Fei, Li}, 155 | booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition}, 156 | pages={4995--5004}, 157 | year={2016} 158 | } 159 | 160 | @article{translation, 161 | title={Neural machine translation by jointly learning to align and translate}, 162 | author={Bahdanau, Dzmitry and Cho, Kyunghyun and Bengio, Yoshua}, 163 | journal={arXiv preprint arXiv:1409.0473}, 164 | year={2014} 165 | } 166 | 167 | @inproceedings{textgen, 168 | title={Generating text with recurrent neural networks}, 169 | author={Sutskever, Ilya and Martens, James and Hinton, Geoffrey E}, 170 | booktitle={Proceedings of the 28th International Conference on Machine Learning (ICML-11)}, 171 | pages={1017--1024}, 172 | year={2011} 173 | } 174 | 175 | @inproceedings{sprec, 176 | title={Deep speech 2: End-to-end speech recognition in english and mandarin}, 177 | author={Amodei, Dario and Ananthanarayanan, Sundaram and Anubhai, Rishita and Bai, Jingliang and Battenberg, Eric and Case, Carl and Casper, Jared and Catanzaro, Bryan and Cheng, Qiang and Chen, Guoliang and others}, 178 | booktitle={International Conference on Machine Learning}, 179 | pages={173--182}, 180 | year={2016} 181 | } 182 | 183 | @article{spsyn, 184 | title={Wavenet: A generative model for raw audio}, 185 | author={Oord, Aaron van den and Dieleman, Sander and Zen, Heiga and Simonyan, Karen and Vinyals, Oriol and Graves, Alex and Kalchbrenner, Nal and Senior, Andrew and Kavukcuoglu, Koray}, 186 | journal={arXiv preprint arXiv:1609.03499}, 187 | year={2016} 188 | } 189 | 190 | @article{atari, 191 | title={Human-level control through deep reinforcement learning}, 192 | author={Mnih, Volodymyr and Kavukcuoglu, Koray and Silver, David and Rusu, Andrei A and Veness, Joel and Bellemare, Marc G and Graves, Alex and Riedmiller, Martin and Fidjeland, Andreas K and Ostrovski, Georg and others}, 193 | journal={Nature}, 194 | volume={518}, 195 | number={7540}, 196 | pages={529--533}, 197 | year={2015}, 198 | publisher={Nature Research} 199 | } 200 | 201 | @article{go, 202 | title={Mastering the game of Go with deep neural networks and tree search}, 203 | author={Silver, David and Huang, Aja and Maddison, Chris J and Guez, Arthur and Sifre, Laurent and Van Den Driessche, George and Schrittwieser, Julian and Antonoglou, Ioannis and Panneershelvam, Veda and Lanctot, Marc and others}, 204 | journal={Nature}, 205 | volume={529}, 206 | number={7587}, 207 | pages={484--489}, 208 | year={2016}, 209 | publisher={Nature Research} 210 | } 211 | 212 | -------------------------------------------------------------------------------- /lecture19/densecap.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/densecap.png -------------------------------------------------------------------------------- /lecture19/lecture19.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/lecture19.pdf -------------------------------------------------------------------------------- /lecture19/loss_functions.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/loss_functions.PNG -------------------------------------------------------------------------------- /lecture19/neural_network.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/neural_network.png -------------------------------------------------------------------------------- /lecture19/poseest.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/poseest.png -------------------------------------------------------------------------------- /lecture19/segmentation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19/segmentation.png -------------------------------------------------------------------------------- /lecture19_new/activation_functions.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19_new/activation_functions.png -------------------------------------------------------------------------------- /lecture19_new/backprop_pic_1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19_new/backprop_pic_1.png -------------------------------------------------------------------------------- /lecture19_new/backprop_pic_2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19_new/backprop_pic_2.png -------------------------------------------------------------------------------- /lecture19_new/gradient_descent_code.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19_new/gradient_descent_code.png -------------------------------------------------------------------------------- /lecture19_new/hello.txt: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /lecture19_new/lecture19.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19_new/lecture19.pdf -------------------------------------------------------------------------------- /lecture19_new/linearly_separable.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19_new/linearly_separable.png -------------------------------------------------------------------------------- /lecture19_new/neural_network.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19_new/neural_network.png -------------------------------------------------------------------------------- /lecture19_new/weights.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture19_new/weights.png -------------------------------------------------------------------------------- /lecture20/Capture.PNG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture20/Capture.PNG -------------------------------------------------------------------------------- /lecture20/Inception.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture20/Inception.png -------------------------------------------------------------------------------- /lecture20/Lecture 20.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture20/Lecture 20.pdf -------------------------------------------------------------------------------- /lecture20/main.tex: -------------------------------------------------------------------------------- 1 | \documentclass{article} 2 | 3 | \usepackage[final]{style} 4 | \usepackage[utf8]{inputenc} % allow utf-8 input 5 | \usepackage[T1]{fontenc} % use 8-bit T1 fonts 6 | \usepackage{hyperref} % hyperlinks 7 | \usepackage{url} % simple URL typesetting 8 | \usepackage{booktabs} % professional-quality tables 9 | \usepackage{amsfonts} % blackboard math symbols 10 | \usepackage{nicefrac} % compact symbols for 1/2, etc. 11 | \usepackage{microtype} % microtypography 12 | \usepackage{verbatim} 13 | \usepackage{graphicx} % for figures 14 | \usepackage[]{algorithm2e} 15 | 16 | 17 | \title{Lecture \#20: Convolutional Neural Networks} 18 | 19 | \author{ 20 | Kendall Beache, Sammy Mohammed, Hannah Zhang \\ 21 | Department of Computer Science\\ 22 | Stanford University\\ 23 | Stanford, CA 94305 \\ 24 | \texttt{\{kbeache, sammym, hzhang16\}@stanford.edu} \\ 25 | } 26 | 27 | \begin{document} 28 | 29 | \maketitle 30 | 31 | \section{Introduction} 32 | Throughout this class, we have explored several different computer vision techniques, including edge detection, clustering methods, classifiers, and feature detectors/descriptors. These techniques still have one main drawback, in that they require humans to hand-design features. In this lecture, we will go over backpropagation in neural networks, a method to recursively find the ideal weights for the neural network, and convolutional neural networks, a new method designed to solve any image processing problem the network is trained for. 33 | 34 | A convolutional neural network is an algorithm that performs \textit{end-to-end learning}, directly mapping raw inputs (images) to a desired output, such as labels or predictions. The history of convolutional neural networks spans decades, but the biggest breakthroughs were in 2010 and 2012, with the publication of Mohamed et al's Acoustic Modeling 35 | using Deep Belief Networks, and Dahel et al's Context-Dependent 36 | Pre-trained Deep Neural 37 | Networks for Large Vocabulary 38 | Speech Recognition. Now, convolutional neural networks and deep learning have led to rapid progress in computer vision, which we will explore. Deep learning approaches usually involve combining a small set of simple tools to build a network and then training that network on data for the specific problem you're trying to solve; they can often be adapted to different problems by simply swapping out the training data. 39 | 40 | \section{Backpropagation in Neural Networks} 41 | 42 | 43 | \subsection{Fundamentals of Backpropagation} 44 | 45 | \textit{Backpropagation} is an algorithm used to build arbitrarily large neural networks, and gradients only require local information to calculate gradients, via recursive implementation of the chain rule. It allows any gradient in the network to be computed by solving the gradients in later layers. Backprop is necessary, as calculating the gradients for intermediate variables manually is difficult even with just a few layers, and does not scale well to neural networks with many layers. 46 | \\ 47 | \begin{figure}[h] 48 | \includegraphics[width=6cm]{Capture.png} 49 | \centering 50 | \caption{1D example (CS 131 lecture slide 20-10)} 51 | \end{figure} 52 | \\ 53 | In order to demonstrate backprop, we will use a 1-d neural network as an example. In this case, $a = wx$, and $\hat{y} =$ max($0,a$). Our goal is to calculate $dL/dW$. Using the chain rule, we can see that $dL/dw_1 = dL/da * da/dw_1$, and $dL/da = dL/d\hat{y} * d\hat{y}/da$. We can then apply the chain rule to solve. 54 | 55 | \subsection{Rules for Calculating Gradients} 56 | 57 | There are some general rules that can help when calculating gradients of input features. If the operation is addition, then the gradients of the input features are distributed in that $\frac{dL}{dx_i} = \frac{dL}{dy}$ for all input features $x_i$. If the operation is multiplication, then each gradient will be proportional to the values of the other input features. For example, if there are two input features $w$ and $x$, then $\frac{dL}{dw} = x\frac{dL}{dy}$ and $\frac{dL}{dx} = w\frac{dL}{dy}$. If the operation is a maximum function, like $y = $ max$(x, w)$, then $\frac{dL}{dw} = \frac{dL}{dy}$ if $w \geq x$ and 0 otherwise; $\frac{dL}{dx} = \frac{dL}{dy}$ if $x \geq w$ and 0 otherwise. If the function is $y = e^x$, then $\frac{dL}{dx} = y \frac{dL}{dy}$, since $\frac{dy}{dx} = e^x$. 58 | 59 | 60 | 61 | \section{Convolutional Neural Networks} 62 | 63 | Convolutional neural networks are a type of deep learning models that are often used for object recognition and classification. At a high level, CNNs start with an original image, and convolve it with multiple filters. The parameters for convolution are also chosen via backpropagation, ensuring that we do not have to hand-choose features. 64 | 65 | Convolutional neural networks can be broken down into 3 steps 66 | 1) Convolution, 2) Non Linearity, 67 | 3) Pooling 68 | 69 | \subsection{Convolution:} We use convolution in this step as a means to extract features from the original input image. A CNN learns the values of the filters or kernels during the training process. Typically we use 1 2D convolution layer on black and white images; however, if you wanted to extract features from a color image you can use 3 2D convolution layers (one for each channel). 70 | 71 | The size of the convolved feature is controlled by three parameters: depth, stride, and zero-padding that we need to decide before the convolution step is performed. Depth corresponds to the number of filters we use for the convolution operation. Stride is the number of pixels by which we slide our filter matrix over the input matrix. We use zero-padding to apply the filter to bordering elements of our input image matrix. Zero padding also allows us to control the size of the feature maps. 72 | 73 | Introduce Non-Linearity (ReLu Function): 74 | After every convolution, the ReLu function is applied to introduce non-linearity in our model (because most real-world data we would want our ConvNet to learn would be non-linear.) ReLU stands for Rectified Linear Unit and its output is given by the following: 75 | \begin{figure}[h] 76 | \includegraphics[width=6cm]{relu.png} 77 | \centering 78 | \caption{The ReLU function. (Kanchan Sarkar, Medium))} 79 | \end{figure} 80 | 81 | Other non-linear functions that can be applied to the convolved images include the sigmoid and tanh function; however, the ReLu function tends to give the best results. 82 | 83 | \subsection{Pooling} 84 | Modern convolutional neural networks incorporate pooling layers -- layers in the neural network that downsize the dimensionality but retain the most important information. This works by defining a spatial neighborhood, such as a 2x2 window, and then taking different data from the neighborhood depending upon the pooling algorithm. One example of this sort of pooling is MaxPool, where the max value in the window is the value that is retained. There are other pooling algorithms, such as AveragePool and SumPool, where the average is taken and the sum of the window is taken. 85 | 86 | \begin{figure}[h] 87 | \includegraphics[width=6cm]{maxpooling.png} 88 | \centering 89 | \caption{Max pooling reduces dimensionality by a factor of 2. (CS 131 lecture slide 20-38)} 90 | \end{figure} 91 | 92 | \subsection{Architecture} 93 | These different layers can be stacked, feeding the input of one layer into the other in order to help improve output quality. These layers extract relevant information, and then classify. 94 | 95 | \begin{figure}[h] 96 | \includegraphics[width=6cm]{structure.png} 97 | \centering 98 | \caption{CNN Architecture (CS 131 lecture slide 20-41)} 99 | \end{figure} 100 | 101 | This basic structure can be expanded -- consider GoogLeNet, a CNN with an Inception layer, a hand-designed network within a network. 102 | 103 | \begin{figure}[h] 104 | \includegraphics[width=6cm]{Inception.png} 105 | \centering 106 | \caption{GoogLeNet architecture and Inception module. (CS 131 lecture slide 19-101)} 107 | \end{figure} 108 | 109 | However, this layering process comes with a drawback. Convolutional neural networks are not shift-invariant, due to the pooling layer -- minor shifts in the image can result in dramatically different classification outputs. (Best demonstrated here: https://richzhang.github.io/antialiased-cnns/). One potential solution would be the BlurPool algorithm, where the window is blurred and shifted before MaxPool is ran, resulting in better outcomes. However, this does not solve the shift-invariance problem, so results may still be unpredictable. 110 | 111 | \section{Conclusion} 112 | While CNNs meet the basic architecture described above, there are still many variations between different networks. CNNs can differ by the amount of layers or values of the hyperparameters used. If you are interested in learning more about different CNNs, read more on GoogLeNet, VGGNet, or ZF Net. 113 | 114 | % References 115 | \small 116 | \bibliographystyle{plain} 117 | \bibliography{bibliography} 118 | \end{document} 119 | -------------------------------------------------------------------------------- /lecture20/maxpooling.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture20/maxpooling.png -------------------------------------------------------------------------------- /lecture20/relu.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture20/relu.png -------------------------------------------------------------------------------- /lecture20/structure.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/StanfordVL/cs131_notes/66bf4b0ad39e245736b9385bed6ecb306ce38fb8/lecture20/structure.png -------------------------------------------------------------------------------- /lecture20/style.sty: -------------------------------------------------------------------------------- 1 | % partial rewrite of the LaTeX2e package for submissions to the 2 | % Conference on Neural Information Processing Systems (NIPS): 3 | % 4 | % - uses more LaTeX conventions 5 | % - line numbers at submission time replaced with aligned numbers from 6 | % lineno package 7 | % - \nipsfinalcopy replaced with [final] package option 8 | % - automatically loads times package for authors 9 | % - loads natbib automatically; this can be suppressed with the 10 | % [nonatbib] package option 11 | % - adds foot line to first page identifying the conference 12 | % 13 | % Roman Garnett (garnett@wustl.edu) and the many authors of 14 | % nips15submit_e.sty, including MK and drstrip@sandia 15 | % 16 | % last revision: August 2016 17 | 18 | \NeedsTeXFormat{LaTeX2e} 19 | \ProvidesPackage{nips_2016}[2016/08/08 NIPS 2016 submission/camera-ready style file] 20 | 21 | % declare final option, which creates camera-ready copy 22 | \newif\if@nipsfinal\@nipsfinalfalse 23 | \DeclareOption{final}{ 24 | \@nipsfinaltrue 25 | } 26 | 27 | % declare nonatbib option, which does not load natbib in case of 28 | % package clash (users can pass options to natbib via 29 | % \PassOptionsToPackage) 30 | \newif\if@natbib\@natbibtrue 31 | \DeclareOption{nonatbib}{ 32 | \@natbibfalse 33 | } 34 | 35 | \ProcessOptions\relax 36 | 37 | % fonts 38 | \renewcommand{\rmdefault}{ptm} 39 | \renewcommand{\sfdefault}{phv} 40 | 41 | % change this every year for notice string at bottom 42 | \newcommand{\@nipsyear}{2017} 43 | \newcommand{\@nipslocation}{Stanford University} 44 | 45 | % handle tweaks for camera-ready copy vs. submission copy 46 | 47 | \newcommand{\@noticestring}{% 48 | Computer Vision: Foundations and Applications (CS 131, \@nipsyear), \@nipslocation.% 49 | } 50 | 51 | % load natbib unless told otherwise 52 | \if@natbib 53 | \RequirePackage{natbib} 54 | \fi 55 | 56 | % set page geometry 57 | \usepackage[verbose=true,letterpaper]{geometry} 58 | \AtBeginDocument{ 59 | \newgeometry{ 60 | textheight=9in, 61 | textwidth=5.5in, 62 | top=1in, 63 | headheight=12pt, 64 | headsep=25pt, 65 | footskip=30pt 66 | } 67 | \@ifpackageloaded{fullpage} 68 | {\PackageWarning{nips_2016}{fullpage package not allowed! Overwriting formatting.}} 69 | {} 70 | } 71 | 72 | \widowpenalty=10000 73 | \clubpenalty=10000 74 | \flushbottom 75 | \sloppy 76 | 77 | % font sizes with reduced leading 78 | \renewcommand{\normalsize}{% 79 | \@setfontsize\normalsize\@xpt\@xipt 80 | \abovedisplayskip 7\p@ \@plus 2\p@ \@minus 5\p@ 81 | \abovedisplayshortskip \z@ \@plus 3\p@ 82 | \belowdisplayskip \abovedisplayskip 83 | \belowdisplayshortskip 4\p@ \@plus 3\p@ \@minus 3\p@ 84 | } 85 | \normalsize 86 | \renewcommand{\small}{% 87 | \@setfontsize\small\@ixpt\@xpt 88 | \abovedisplayskip 6\p@ \@plus 1.5\p@ \@minus 4\p@ 89 | \abovedisplayshortskip \z@ \@plus 2\p@ 90 | \belowdisplayskip \abovedisplayskip 91 | \belowdisplayshortskip 3\p@ \@plus 2\p@ \@minus 2\p@ 92 | } 93 | \renewcommand{\footnotesize}{\@setfontsize\footnotesize\@ixpt\@xpt} 94 | \renewcommand{\scriptsize}{\@setfontsize\scriptsize\@viipt\@viiipt} 95 | \renewcommand{\tiny}{\@setfontsize\tiny\@vipt\@viipt} 96 | \renewcommand{\large}{\@setfontsize\large\@xiipt{14}} 97 | \renewcommand{\Large}{\@setfontsize\Large\@xivpt{16}} 98 | \renewcommand{\LARGE}{\@setfontsize\LARGE\@xviipt{20}} 99 | \renewcommand{\huge}{\@setfontsize\huge\@xxpt{23}} 100 | \renewcommand{\Huge}{\@setfontsize\Huge\@xxvpt{28}} 101 | 102 | % sections with less space 103 | \providecommand{\section}{} 104 | \renewcommand{\section}{% 105 | \@startsection{section}{1}{\z@}% 106 | {-2.0ex \@plus -0.5ex \@minus -0.2ex}% 107 | { 1.5ex \@plus 0.3ex \@minus 0.2ex}% 108 | {\large\bf\raggedright}% 109 | } 110 | \providecommand{\subsection}{} 111 | \renewcommand{\subsection}{% 112 | \@startsection{subsection}{2}{\z@}% 113 | {-1.8ex \@plus -0.5ex \@minus -0.2ex}% 114 | { 0.8ex \@plus 0.2ex}% 115 | {\normalsize\bf\raggedright}% 116 | } 117 | \providecommand{\subsubsection}{} 118 | \renewcommand{\subsubsection}{% 119 | \@startsection{subsubsection}{3}{\z@}% 120 | {-1.5ex \@plus -0.5ex \@minus -0.2ex}% 121 | { 0.5ex \@plus 0.2ex}% 122 | {\normalsize\bf\raggedright}% 123 | } 124 | \providecommand{\paragraph}{} 125 | \renewcommand{\paragraph}{% 126 | \@startsection{paragraph}{4}{\z@}% 127 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 128 | {-1em}% 129 | {\normalsize\bf}% 130 | } 131 | \providecommand{\subparagraph}{} 132 | \renewcommand{\subparagraph}{% 133 | \@startsection{subparagraph}{5}{\z@}% 134 | {1.5ex \@plus 0.5ex \@minus 0.2ex}% 135 | {-1em}% 136 | {\normalsize\bf}% 137 | } 138 | \providecommand{\subsubsubsection}{} 139 | \renewcommand{\subsubsubsection}{% 140 | \vskip5pt{\noindent\normalsize\rm\raggedright}% 141 | } 142 | 143 | % float placement 144 | \renewcommand{\topfraction }{0.85} 145 | \renewcommand{\bottomfraction }{0.4} 146 | \renewcommand{\textfraction }{0.1} 147 | \renewcommand{\floatpagefraction}{0.7} 148 | 149 | \newlength{\@nipsabovecaptionskip}\setlength{\@nipsabovecaptionskip}{7\p@} 150 | \newlength{\@nipsbelowcaptionskip}\setlength{\@nipsbelowcaptionskip}{\z@} 151 | 152 | \setlength{\abovecaptionskip}{\@nipsabovecaptionskip} 153 | \setlength{\belowcaptionskip}{\@nipsbelowcaptionskip} 154 | 155 | % swap above/belowcaptionskip lengths for tables 156 | \renewenvironment{table} 157 | {\setlength{\abovecaptionskip}{\@nipsbelowcaptionskip}% 158 | \setlength{\belowcaptionskip}{\@nipsabovecaptionskip}% 159 | \@float{table}} 160 | {\end@float} 161 | 162 | % footnote formatting 163 | \setlength{\footnotesep }{6.65\p@} 164 | \setlength{\skip\footins}{9\p@ \@plus 4\p@ \@minus 2\p@} 165 | \renewcommand{\footnoterule}{\kern-3\p@ \hrule width 12pc \kern 2.6\p@} 166 | \setcounter{footnote}{0} 167 | 168 | % paragraph formatting 169 | \setlength{\parindent}{\z@} 170 | \setlength{\parskip }{5.5\p@} 171 | 172 | % list formatting 173 | \setlength{\topsep }{4\p@ \@plus 1\p@ \@minus 2\p@} 174 | \setlength{\partopsep }{1\p@ \@plus 0.5\p@ \@minus 0.5\p@} 175 | \setlength{\itemsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 176 | \setlength{\parsep }{2\p@ \@plus 1\p@ \@minus 0.5\p@} 177 | \setlength{\leftmargin }{3pc} 178 | \setlength{\leftmargini }{\leftmargin} 179 | \setlength{\leftmarginii }{2em} 180 | \setlength{\leftmarginiii}{1.5em} 181 | \setlength{\leftmarginiv }{1.0em} 182 | \setlength{\leftmarginv }{0.5em} 183 | \def\@listi {\leftmargin\leftmargini} 184 | \def\@listii {\leftmargin\leftmarginii 185 | \labelwidth\leftmarginii 186 | \advance\labelwidth-\labelsep 187 | \topsep 2\p@ \@plus 1\p@ \@minus 0.5\p@ 188 | \parsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 189 | \itemsep \parsep} 190 | \def\@listiii{\leftmargin\leftmarginiii 191 | \labelwidth\leftmarginiii 192 | \advance\labelwidth-\labelsep 193 | \topsep 1\p@ \@plus 0.5\p@ \@minus 0.5\p@ 194 | \parsep \z@ 195 | \partopsep 0.5\p@ \@plus 0\p@ \@minus 0.5\p@ 196 | \itemsep \topsep} 197 | \def\@listiv {\leftmargin\leftmarginiv 198 | \labelwidth\leftmarginiv 199 | \advance\labelwidth-\labelsep} 200 | \def\@listv {\leftmargin\leftmarginv 201 | \labelwidth\leftmarginv 202 | \advance\labelwidth-\labelsep} 203 | \def\@listvi {\leftmargin\leftmarginvi 204 | \labelwidth\leftmarginvi 205 | \advance\labelwidth-\labelsep} 206 | 207 | % create title 208 | \providecommand{\maketitle}{} 209 | \renewcommand{\maketitle}{% 210 | \par 211 | \begingroup 212 | \renewcommand{\thefootnote}{\fnsymbol{footnote}} 213 | % for perfect author name centering 214 | \renewcommand{\@makefnmark}{\hbox to \z@{$^{\@thefnmark}$\hss}} 215 | % The footnote-mark was overlapping the footnote-text, 216 | % added the following to fix this problem (MK) 217 | \long\def\@makefntext##1{% 218 | \parindent 1em\noindent 219 | \hbox to 1.8em{\hss $\m@th ^{\@thefnmark}$}##1 220 | } 221 | \thispagestyle{empty} 222 | \@maketitle 223 | \@thanks 224 | \@notice 225 | \endgroup 226 | \let\maketitle\relax 227 | \let\thanks\relax 228 | } 229 | 230 | % rules for title box at top of first page 231 | \newcommand{\@toptitlebar}{ 232 | \hrule height 4\p@ 233 | \vskip 0.25in 234 | \vskip -\parskip% 235 | } 236 | \newcommand{\@bottomtitlebar}{ 237 | \vskip 0.29in 238 | \vskip -\parskip 239 | \hrule height 1\p@ 240 | \vskip 0.09in% 241 | } 242 | 243 | % create title (includes both anonymized and non-anonymized versions) 244 | \providecommand{\@maketitle}{} 245 | \renewcommand{\@maketitle}{% 246 | \vbox{% 247 | \hsize\textwidth 248 | \linewidth\hsize 249 | \vskip 0.1in 250 | \@toptitlebar 251 | \centering 252 | {\LARGE\bf \@title\par} 253 | \@bottomtitlebar 254 | \if@nipsfinal 255 | \def\And{% 256 | \end{tabular}\hfil\linebreak[0]\hfil% 257 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 258 | } 259 | \def\AND{% 260 | \end{tabular}\hfil\linebreak[4]\hfil% 261 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\ignorespaces% 262 | } 263 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@}\@author\end{tabular}% 264 | \else 265 | \begin{tabular}[t]{c}\bf\rule{\z@}{24\p@} 266 | Anonymous Author(s) \\ 267 | Affiliation \\ 268 | Address \\ 269 | \texttt{email} \\ 270 | \end{tabular}% 271 | \fi 272 | \vskip 0.3in \@minus 0.1in 273 | } 274 | } 275 | 276 | % add conference notice to bottom of first page 277 | \newcommand{\ftype@noticebox}{8} 278 | \newcommand{\@notice}{% 279 | % give a bit of extra room back to authors on first page 280 | \enlargethispage{2\baselineskip}% 281 | \@float{noticebox}[b]% 282 | \footnotesize\@noticestring% 283 | \end@float% 284 | } 285 | 286 | % abstract styling 287 | \renewenvironment{abstract}% 288 | {% 289 | \vskip 0.075in% 290 | \centerline% 291 | {\large\bf Abstract}% 292 | \vspace{0.5ex}% 293 | \begin{quote}% 294 | } 295 | { 296 | \par% 297 | \end{quote}% 298 | \vskip 1ex% 299 | } 300 | 301 | \endinput -------------------------------------------------------------------------------- /template/bibliography.bib: -------------------------------------------------------------------------------- 1 | @article{krishna2017visual, 2 | title={Visual genome: Connecting language and vision using crowdsourced dense image annotations}, 3 | author={Krishna, Ranjay and Zhu, Yuke and Groth, Oliver and Johnson, Justin and Hata, Kenji and Kravitz, Joshua and Chen, Stephanie and Kalantidis, Yannis and Li, Li-Jia and Shamma, David A and others}, 4 | journal={International Journal of Computer Vision}, 5 | volume={123}, 6 | number={1}, 7 | pages={32--73}, 8 | year={2017}, 9 | publisher={Springer} 10 | } 11 | -------------------------------------------------------------------------------- /template/template.tex: -------------------------------------------------------------------------------- 1 | \documentclass{article} 2 | 3 | \usepackage[final]{style} 4 | \usepackage[utf8]{inputenc} % allow utf-8 input 5 | \usepackage[T1]{fontenc} % use 8-bit T1 fonts 6 | \usepackage{hyperref} % hyperlinks 7 | \usepackage{url} % simple URL typesetting 8 | \usepackage{booktabs} % professional-quality tables 9 | \usepackage{amsfonts} % blackboard math symbols 10 | \usepackage{nicefrac} % compact symbols for 1/2, etc. 11 | \usepackage{microtype} % microtypography 12 | \usepackage{verbatim} 13 | \usepackage{graphicx} % for figures 14 | 15 | \title{Lecture \#X: NAME OF LECTURE} 16 | 17 | \author{ 18 | Student1 name, student2 name, etc. \\ 19 | Department of Computer Science\\ 20 | Stanford University\\ 21 | Stanford, CA 94305 \\ 22 | \texttt{\{STUDENT1, STUDENT2, etc.\}@cs.stanford.edu} \\ 23 | } 24 | 25 | \begin{document} 26 | 27 | \maketitle 28 | 29 | 30 | \section{Introduction} 31 | Copy this template and use it to write up the lecture notes. Also copy over 32 | bibliography.bib and add any references you use to that file. 33 | 34 | % References 35 | \small 36 | \bibliographystyle{plain} 37 | \bibliography{bibliography} 38 | \end{document} 39 | --------------------------------------------------------------------------------