├── README.md
├── index.html
├── index.php
├── styles.css
└── upload.php
/README.md:
--------------------------------------------------------------------------------
1 |
2 | # Tensorflow-Image-Classifier-Web-Demo
3 | "Calorie Counter", helps you stay fit , by using computer vision to find which item are you eating and giving you nutrition facts
4 |
5 | ###End result (It's a video , click it)
6 | [](http://www.youtube.com/watch?v=ECsLu60n0ss)
7 |
8 |
9 | ### Requirements
10 |
11 | Web Server of your choice (nginx/apache/etc)(PHP must be installed)(See Any tutorial for setting up a webserver)
12 |
13 |
14 |
15 | ### Files, What do they do?
16 | uploads = user submitted images go here
17 |
18 | index.html = upload site
19 |
20 | postdata.php = main page, processing
21 |
22 | styles.css = CSS styles for the web page
23 |
24 |
25 |
26 | ###Inspiration:
27 | 
28 | Source: SAO Ordinal Scale - A1 Pictures
29 |
30 | ###First Step
31 | Clone this project (In terminal)
32 | ```
33 | git clone https://github.com/AxelAli/Tensorflow-Image-Classifier-Web-Demo.git
34 | cd Tensorflow-Image-Classifier-Web-Demo
35 | ```
36 | then
37 |
38 | ###Setting a tensorflow image recognizer
39 | Follow this project.
40 | ```
41 | https://github.com/AxelAli/Tensorflow-Image-Classification
42 | ```
43 | When finished
44 | ```
45 | cd ..
46 | ```
47 |
48 |
49 | (We should now be in this proyect folder)
50 |
51 |
52 | ###Usage
53 | Open your webserver
54 | ####Click Choose file (in mobile you just open the camera)
55 | 
56 |
57 | ####Select your picture
58 | 
59 |
60 | ####Click Upload
61 | 
62 |
63 | ####Results are in!
64 | 
65 |
66 | #####Orange is the image that you uploaded
67 | #####Purple is the name of the food and serving size
68 | #####Green is the amounts of each item
69 |
70 |
71 | ### Known Bugs / Errors / Improvements
72 | >This should be a single webpage
73 |
74 | >if it doesnt find anything it shows a random item (we should check Accuracy of <80%)
75 |
76 | >data is hardcoded, this should be made with a database
77 |
78 | >if it doenst know which item it is, it should ask the user what is it and run the code to add it automatically
79 |
--------------------------------------------------------------------------------
/index.html:
--------------------------------------------------------------------------------
1 |
2 |