├── README.md ├── 顶会红外相关.md └── 红外检测论文汇总.md /README.md: -------------------------------------------------------------------------------- 1 | # Thermal-infrared-object-detection 2 | a simple repo for some references about thermal infrared object detection 3 | -------------------------------------------------------------------------------- /顶会红外相关.md: -------------------------------------------------------------------------------- 1 | # 顶会红外相关 2 | 3 | ## 1. CVPR 2021 4 | 5 | - Neural Feature Search for RGB-Infrared Person Re-Identification:heavy_check_mark:(未完成ppt) 6 | - Discover Cross-Modality Nuances for Visible-Infrared Person Re-Identification:heavy_check_mark:(未完成ppt) 7 | - ABMDRNet_ Adaptive-Weighted Bi-Directional Modality Difference Reduction Network for RGB-T Semantic Segmentation:heavy_check_mark:(未完成ppt) 8 | - Cross-Modal Collaborative Representation Learning and a Large-Scale RGBT Benchmark for Crowd Counting:heavy_check_mark:(未完成ppt) 9 | - There Is More Than Meets the Eye_ Self-Supervised Multi-Object Detection and Tracking With Sound by Distilling Multimodal Knowledge:heavy_check_mark: 10 | 11 | ## 2. CVPR 2020 12 | 13 | - Cross-Modal Pattern-Propagation for RGB-T Tracking:heavy_check_mark:(未完成ppt，有初稿) 14 | - Hi-CMD: Hierarchical Cross-Modality Disentanglement for Visible-Infrared Person Re-Identification:heavy_check_mark:(未完成ppt，有初稿) 15 | 16 | ## 3. CVPR 2019 17 | 18 | - Learning to Reduce Dual-Level Discrepancy for Infrared-Visible Person Re-Identification:heavy_check_mark: 19 | - ContactDB: Analyzing and Predicting Grasp Contact via Thermal Imaging:heavy_check_mark:(未完成ppt，有初稿) 20 | - Time-Resolved Light Transport Decomposition for Thermal Photometric Stereo:heavy_check_mark: 21 | 22 | ## 4. ICCV 2021 23 | 24 | - Channel Augmented Joint Learning for Visible-Infrared Recognition:heavy_check_mark:(未完成ppt) 25 | - CM-NAS: Cross-Modality Neural Architecture Search for Visible-Infrared Person Re-Identification:heavy_check_mark:(未完成ppt) 26 | - Syncretic Modality Collaborative Learning for Visible Infrared Person Re-Identification:heavy_check_mark:(未完成ppt) 27 | - Learning by Aligning: Visible-Infrared Person Re-Identification Using Cross-Modal Correspondences:heavy_check_mark:(未完成ppt) 28 | - Robust Small-scale Pedestrian Detection with Cued Recall via Memory Learning:heavy_check_mark: 29 | 30 | ## 5. ICCV 2019 31 | 32 | - Non-Local Intrinsic Decomposition With Near-Infrared Priors:heavy_check_mark:（理解不够透彻） 33 | - RGB-Infrared Cross-Modality Person Re-Identification via Joint Pixel and Feature Alignment:heavy_check_mark: 34 | - Miss Detection vs. False Alarm: Adversarial Learning for Small Object Segmentation in Infrared Images:heavy_check_mark: 35 | 36 | ## 6. ECCV 2020 37 | 38 | - Challenge-Aware RGBT Tracking:heavy_check_mark: 39 | - Task-conditioned Domain Adaptation for Pedestrian Detection in Thermal Imagery:heavy_check_mark: 40 | - Dynamic Dual-Attentive Aggregation Learning for Visible-Infrared Person Re-Identification:heavy_check_mark: 41 | - Lensless Imaging with Focusing Sparse URA Masks in Long-Wave Infrared and its Application for Human Detection:heavy_exclamation_mark:(半成品，读到一半读不下去了) 42 | 43 | 共计22篇文章，其中10篇行人重识别，3篇RGB-T跟踪，2篇行人检测，1篇RGB-T语义分割，1篇小目标检测，1篇人群计数，1篇成像优化，1篇近红外先验的理论研究，1篇抓取接触预测，1篇红外光度立体测量。 44 | 45 | ## 7. CVPR 2022 46 | 47 | - Infrared Invisible Clothing: Hiding from Infrared Detectors at Multiple Angles in Real World (用特种条纹干扰红外检测器) 48 | - Visible-Thermal UAV Tracking: A Large-Scale Benchmark and New Baseline （数据集-双光UAV跟踪） 49 | - ISNet: Shape Matters for Infrared Small Target Detection （红外小目标检测） 50 | - FMCNet: Feature-Level Modality Compensation for Visible-Infrared Person Re-Identification （双光Re-id） 51 | - Learning with Twin Noisy Labels for Visible-Infrared Person Re-Identification （双光Re-id） 52 | - Shape from Thermal Radiation: Passive Ranging Using Multi-spectral LWIR Measurements （解析影响红外成像的各种因素并还原出深度图） 53 | - Target-aware Dual Adversarial Learning and a Multi-scenario Multi-Modality Benchmark to Fuse Infrared and Visible for Object Detection （双光目标检测） 54 | - Learning Modal-Invariant and Temporal-Memory for Video-based Visible-Infrared Person Re-Identification （双光Re-id） 55 | 56 | ## 8. WACV 2022 57 | 58 | - Meta-UDA: Unsupervised Domain Adaptive Thermal Object Detection using Meta-Learning （基于元学习的无监督域适应热红外目标检测） -------------------------------------------------------------------------------- /红外检测论文汇总.md: -------------------------------------------------------------------------------- 1 | # 红外论文汇总 2 | 3 | # 0. 符号与图表 4 | 5 | CCF评级： 6 | 7 | CCF A 8 | 9 | CCF B 10 | 11 | CCF C 12 | 13 | CCF None 14 | 15 | # 1. Unsupervised thermal-to-visible domain adaptation method for pedestrian detection CCF C 16 | 17 | > ```bibtex 18 | > @article{marnissi2022unsupervised, 19 | > title={Unsupervised thermal-to-visible domain adaptation method for pedestrian detection}, 20 | > author={Marnissi, Mohamed Amine and Fradi, Hajer and Sahbani, Anis and Amara, Najoua Essoukri Ben}, 21 | > journal={Pattern Recognition Letters}, 22 | > volume={153}, 23 | > pages={222--231}, 24 | > year={2022}, 25 | > publisher={Elsevier} 26 | > } 27 | > ``` 28 | 29 | ![image-20220601171322150](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206011713273.png) 30 | 31 | 32 | 33 | 主要解决标注不易获得问题。 34 | 35 | # 2. Unsupervised image-generation enhanced adaptation for object detection in thermal images CCF None 36 | 37 | > ``` 38 | > @article{liu2020unsupervised, 39 | > title={Unsupervised image-generation enhanced adaptation for object detection in thermal images}, 40 | > author={Liu, Peng and Li, Fuyu and Yuan, Shanshan and Li, Wanyi}, 41 | > journal={Mobile Information Systems}, 42 | > volume={2021}, 43 | > year={2020}, 44 | > publisher={Hindawi} 45 | > } 46 | > ``` 47 | 48 | ![image-20220526143715339](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526143715339.png?token=AJU4EWOUJJGK3WLOBTPSNNLCR4QFW) 49 | 50 | ![image-20220526143735186](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526143735186.png?token=AJU4EWO4S6QOON5M5IKL6T3CR4QG4) 51 | 52 | 在CycleGAN的基础上提出了一种无监督图像生成的域适应增强方法来提高热红外检测器性能。 53 | 54 | # 3. Thermal object detection using domain adaptation through style consistency CCF None 55 | 56 | > ``` 57 | > @article{munir2020thermal, 58 | > title={Thermal object detection using domain adaptation through style consistency}, 59 | > author={Munir, Farzeen and Azam, Shoaib and Rafique, Muhammad Aasim and Sheri, Ahmad Muqeem and Jeon, Moongu}, 60 | > year={2020} 61 | > } 62 | > ``` 63 | 64 | ![image-20220526144149686](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526144149686.png?token=AJU4EWKHWRZVRZNTKRRDT3TCR4QW2) 65 | 66 | ![image-20220526144214953](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526144214953.png?token=AJU4EWJMFG7Q6LPSWUA7D2TCR4QYM) 67 | 68 | ![image-20220526144236886](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526144236886.png?token=AJU4EWIBZYEPLEFMK7NINFLCR4QZ2) 69 | 70 | 提出了一种在红外检测中的域适应方法。论文探索了多种域适应方法。首先，使用GAN利用Style consistency将low-level的特征从可见光谱转移到红外光谱。其次，利用带有Style consistency的cross-domain的模块将训练好的可见光检测器迁移到红外域中。这里提到了The multi-style transfer is applied to transfer the low-level features such as curvatures and edges from the source domain to the target domain. 也就是举例指出所谓low-level的特征包括但不限于曲率和边缘。 71 | 72 | # 4. Thermal Image Enhancement using Generative Adversarial Network for Pedestrian Detection CCF C 73 | 74 | > ``` 75 | > @inproceedings{marnissi2021thermal, 76 | > title={Thermal Image Enhancement using Generative Adversarial Network for Pedestrian Detection}, 77 | > author={Marnissi, Mohamed Amine and Fradi, Hajer and Sahbani, Anis and Amara, Najoua Essoukri Ben}, 78 | > booktitle={2020 25th International Conference on Pattern Recognition (ICPR)}, 79 | > pages={6509--6516}, 80 | > year={2021}, 81 | > organization={IEEE} 82 | > } 83 | > https://github.com/AmineMarnissi/TE-GAN 84 | > ``` 85 | 86 | ![image-20220526151358035](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526151358035.png?token=AJU4EWOXGFQVZVNNSZOPHYLCR4UPM) 87 | 88 | 提出了对比增强模块和去噪模块，引入了边缘恢复的步骤。这些不同的线索可以融合到同一个框架中。 89 | 90 | # 5. Task-conditioned Domain Adaptation for Pedestrian Detection in Thermal Imagery CCF B 91 | 92 | > ``` 93 | > @inproceedings{kieu2020task, 94 | > title={Task-conditioned domain adaptation for pedestrian detection in thermal imagery}, 95 | > author={Kieu, My and Bagdanov, Andrew D and Bertini, Marco and Bimbo, Alberto del}, 96 | > booktitle={European Conference on Computer Vision}, 97 | > pages={546--562}, 98 | > year={2020}, 99 | > organization={Springer} 100 | > } 101 | > ``` 102 | 103 | ![image-20220526152344776](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526152344776.png?token=AJU4EWK67U5PWFJMC4H2NR3CR4VT6) 104 | 105 | ![image-20220526152401158](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526152401158.png?token=AJU4EWKPXUEVRO2V7YKZGI3CR4VVA) 106 | 107 | ![image-20220526152415148](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526152415148.png?token=AJU4EWKQYEP6KN4ELAB3NZDCR4VV4) 108 | 109 | ![image-20220526152437319](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526152437319.png?token=AJU4EWOVUFOXTRDFKLKOY3TCR4VXI) 110 | 111 | ![image-20220526152454310](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526152454310.png?token=AJU4EWO3FJBQBK6AD2CDYELCR4VYM) 112 | 113 | 通过引入一个辅助分类网络，判断图像是白天还是夜间，去调整输出的特征，帮助将yolov3适应到红外域中。 114 | 115 | # 6. SSTN: Self-Supervised Domain Adaptation Thermal Object Detection for Autonomous Driving CCF C 116 | 117 | > ``` 118 | > @inproceedings{munir2021sstn, 119 | > title={Sstn: Self-supervised domain adaptation thermal object detection for autonomous driving}, 120 | > author={Munir, Farzeen and Azam, Shoaib and Jeon, Moongu}, 121 | > booktitle={2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)}, 122 | > pages={206--213}, 123 | > year={2021}, 124 | > organization={IEEE} 125 | > } 126 | > ``` 127 | 128 | ![image-20220526153310575](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526153310575.png?token=AJU4EWLP4X3ZZOVLFMSKW3DCR4WXK) 129 | 130 | ![image-20220526153332888](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526153332888.png?token=AJU4EWMXZKJN4YQSGY2QIG3CR4WYY) 131 | 132 | ![image-20220526153403403](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526153403403.png?token=AJU4EWLVWVH4TKZMTH4HMA3CR4W2U) 133 | 134 | ![image-20220526153414865](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526153414865.png?token=AJU4EWPC4WNEIM4NHF6SPHDCR4W3K) 135 | 136 | 对比学习+transformer解决红外检测问题。模型架构较为简单，值得仔细看看是怎么描述方法的，毕竟A+B 137 | 138 | # 7. Spatio-Contextual Deep Network Based Multimodal Pedestrian Detection For Autonomous Driving CCF B :star::star::star::star::star: 139 | 140 | > ``` 141 | > @article{dasgupta2022spatio, 142 | > title={Spatio-contextual deep network-based multimodal pedestrian detection for autonomous driving}, 143 | > author={Dasgupta, Kinjal and Das, Arindam and Das, Sudip and Bhattacharya, Ujjwal and Yogamani, Senthil}, 144 | > journal={IEEE Transactions on Intelligent Transportation Systems}, 145 | > year={2022}, 146 | > publisher={IEEE} 147 | > } 148 | > ``` 149 | 150 | ![image-20220526155604677](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526155604677.png?token=AJU4EWJT7MJHTAUWVMVPRB3CR4ZNI) 151 | 152 | ![image-20220526155722118](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526155722118.png?token=AJU4EWNYDON74PXF2NG6NT3CR4ZSC) 153 | 154 | ![image-20220526155738040](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526155738040.png?token=AJU4EWOX6SYK66GXYSE3R43CR4ZTA) 155 | 156 | 文章在Introduction第二段提到： 157 | 158 | > The infrared spectrum lies below the visible spectrum used 159 | > by RGB cameras. It is categorized into near and far-infrared 160 | > regions with a region in the middle, further sub-divided into 161 | > short, mid, and long wavelengths. Mid and long-wavelength 162 | > infrared are commonly known as thermal infrared as the intensity 163 | > of captured radiation are correlated with the temperature 164 | > of an object [9]. 165 | 166 | 对红外图像做了区分，可以跟我文章中的互相佐证。相关工作介绍了Efficient pedestrian detection at nighttime using a thermal camera，这是我之前没有查到的文章。值得好好看一下 167 | 168 | # 8. Robust Small-scale Pedestrian Detection with Cued Recall via Memory Learning CCF A 169 | 170 | > ``` 171 | > @inproceedings{kim2021robust, 172 | > title={Robust small-scale pedestrian detection with cued recall via memory learning}, 173 | > author={Kim, Jung Uk and Park, Sungjune and Ro, Yong Man}, 174 | > booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision}, 175 | > pages={3050--3059}, 176 | > year={2021} 177 | > } 178 | > ``` 179 | 180 | ![image-20220526163309947](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526163309947.png?token=AJU4EWP7QJYYMFKIA5RZP4LCR45YK) 181 | 182 | ![image-20220526163331314](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526163331314.png?token=AJU4EWIENKPFYOJGNHQ435TCR45ZU) 183 | 184 | 通过一个memory bank把不同尺度的特征存储起来，再将样本图片提取的行人特征根据大小分成不同尺度，对不同尺度分别做不同操作。 185 | 186 | # 9. Robust pedestrian detection in thermal imagery using synthesized images CCF C 187 | 188 | > ``` 189 | > @inproceedings{kieu2021robust, 190 | > title={Robust pedestrian detection in thermal imagery using synthesized images}, 191 | > author={Kieu, My and Berlincioni, Lorenzo and Galteri, Leonardo and Bertini, Marco and Bagdanov, Andrew D and Del Bimbo, Alberto}, 192 | > booktitle={2020 25th International Conference on Pattern Recognition (ICPR)}, 193 | > pages={8804--8811}, 194 | > year={2021}, 195 | > organization={IEEE} 196 | > } 197 | > ``` 198 | 199 | ![image-20220526164625130](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526164625130.png?token=AJU4EWMGJGWLAM6YITPRHMTCR47KA) 200 | 201 | 利用LS-GAN网络生成虚假的红外图像，并将虚假红外图像与真是红外图像混合做为一种数据增强的方式训练检测器 202 | 203 | # 10. Pedestrian Detection in Thermal Images using Saliency Maps （CVPR Workshop） 204 | 205 | > ``` 206 | > @inproceedings{ghose2019pedestrian, 207 | > title={Pedestrian detection in thermal images using saliency maps}, 208 | > author={Ghose, Debasmita and Desai, Shasvat M and Bhattacharya, Sneha and Chakraborty, Deep and Fiterau, Madalina and Rahman, Tauhidur}, 209 | > booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops}, 210 | > pages={0--0}, 211 | > year={2019} 212 | > } 213 | > ``` 214 | 215 | ![image-20220526165608426](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526165608426.png?token=AJU4EWKBPYLNCVXFVEKBUADCR5AOQ) 216 | 217 | 使用红外图像的显著图做为注意力机制增强原有的红外图像，以提高图像对比度，解决白天的热红外图像由于对比度降低导致的检测器性能下降问题。 218 | 219 | # 11. Pedestrian detection from thermal images: A sparse representation based approach CCF None 220 | 221 | > ``` 222 | > @article{qi2016pedestrian, 223 | > title={Pedestrian detection from thermal images: A sparse representation based approach}, 224 | > author={Qi, Bin and John, Vijay and Liu, Zheng and Mita, Seiichi}, 225 | > journal={Infrared Physics \& Technology}, 226 | > volume={76}, 227 | > pages={157--167}, 228 | > year={2016}, 229 | > publisher={Elsevier} 230 | > } 231 | > ``` 232 | 233 | ![image-20220526170743498](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526170743498.png?token=AJU4EWMP5GXWW6ROPF5BM23CR5BZ6) 234 | 235 | ![image-20220526170810214](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526170810214.png?token=AJU4EWPJQ5X4FXY7VMM442TCR5B3S) 236 | 237 | ![image-20220526170826155](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526170826155.png?token=AJU4EWOTRMIFE4OKTOMAR4DCR5B4S) 238 | 239 | 2016年的工作，是一个比较上古的文章 240 | 241 | # 12. Pedestrian Detection Based on Center, Temperature, Scale and Ratio Prediction in Thermal Imagery CCF None 242 | 243 | > ``` 244 | > @inproceedings{lu2021pedestrian, 245 | > title={Pedestrian Detection Based on Center, Temperature, Scale and Ratio Prediction in Thermal Imagery}, 246 | > author={Lu, Chen and Zhang, Senlin and Liu, Meiqin}, 247 | > booktitle={2021 40th Chinese Control Conference (CCC)}, 248 | > pages={7288--7293}, 249 | > year={2021}, 250 | > organization={IEEE} 251 | > } 252 | > ``` 253 | 254 | ![image-20220526170714121](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526170714121.png?token=AJU4EWJFLWPOLCKTE3IYP5DCR5BYC) 255 | 256 | 文章创新性存疑 257 | 258 | # 13. Pedestrian Detection and Tracking from Low-Resolution Unmanned Aerial Vehicle Thermal Imagery CCF None 259 | 260 | > ``` 261 | > @article{ma2016pedestrian, 262 | > title={Pedestrian detection and tracking from low-resolution unmanned aerial vehicle thermal imagery}, 263 | > author={Ma, Yalong and Wu, Xinkai and Yu, Guizhen and Xu, Yongzheng and Wang, Yunpeng}, 264 | > journal={Sensors}, 265 | > volume={16}, 266 | > number={4}, 267 | > pages={446}, 268 | > year={2016}, 269 | > publisher={Multidisciplinary Digital Publishing Institute} 270 | > } 271 | > ``` 272 | 273 | ![image-20220526171951089](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526171951089.png?token=AJU4EWKYUV3FWG7XIC52AD3CR5DHM) 274 | 275 | 也是一篇比较早的文章，主要是做无人机视角下的红外检测与跟踪。提出了一个数据集，但是没有开源 276 | 277 | # 14. Partially fake it till you make it: mixing real and fake thermal images for improved object detection CCF A 278 | 279 | > ``` 280 | > @inproceedings{bongini2021partially, 281 | > title={Partially fake it till you make it: mixing real and fake thermal images for improved object detection}, 282 | > author={Bongini, Francesco and Berlincioni, Lorenzo and Bertini, Marco and Del Bimbo, Alberto}, 283 | > booktitle={Proceedings of the 29th ACM International Conference on Multimedia}, 284 | > pages={5482--5490}, 285 | > year={2021} 286 | > } 287 | > ``` 288 | 289 | ![image-20220526191640177](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526191640177.png?token=AJU4EWO4C6L4ZBUP5AR7OSLCR5RAI) 290 | 291 | 提出了一系列数据增强的方法用来拓展红外数据集的规模，生成一系列真假混合的红外图像。这些真假混合的图像是用3D物体模型放置到图片上后通过GAN网络再转换到与红外图像风格一致。 292 | 293 | # 15. Object Detection on Thermal Images for Unmanned Aerial Vehicles Using Domain Adaption Through Fine-Tuning CCF None 294 | 295 | > ``` 296 | > @inproceedings{rauch2021object, 297 | > title={Object Detection on Thermal Images for Unmanned Aerial Vehicles Using Domain Adaption Through Fine-Tuning}, 298 | > author={Rauch, Jonas and Doer, Christopher and Trommer, Gert F}, 299 | > booktitle={2021 28th Saint Petersburg International Conference on Integrated Navigation Systems (ICINS)}, 300 | > pages={1--4}, 301 | > year={2021}, 302 | > organization={IEEE} 303 | > } 304 | > ``` 305 | 306 | ![image-20220526194501603](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526194501603.png?token=AJU4EWNF5MU4HODESYLGBV3CR5UHY) 307 | 308 | 没看，写的看不懂，大概贡献可能就是把检测算法嵌入到计算资源受限的无人机芯片上了，但是文章本身写的太难理解了。 309 | 310 | # 16. Multispectral Pedestrian Detection: Benchmark Dataset and Baseline CCF A 311 | 312 | > ``` 313 | > @inproceedings{hwang2015multispectral, 314 | > title={Multispectral pedestrian detection: Benchmark dataset and baseline}, 315 | > author={Hwang, Soonmin and Park, Jaesik and Kim, Namil and Choi, Yukyung and So Kweon, In}, 316 | > booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition}, 317 | > pages={1037--1045}, 318 | > year={2015} 319 | > } 320 | > ``` 321 | 322 | KAIST 数据集论文 323 | 324 | # 17. Multispectral Pedestrian Detection via Simultaneous Detection and Segmentation CCF C 325 | 326 | > ``` 327 | > @article{li2018multispectral, 328 | > title={Multispectral pedestrian detection via simultaneous detection and segmentation}, 329 | > author={Li, Chengyang and Song, Dan and Tong, Ruofeng and Tang, Min}, 330 | > journal={arXiv preprint arXiv:1808.04818}, 331 | > year={2018} 332 | > } 333 | > ``` 334 | 335 | ![image-20220526195006720](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526195006720.png?token=AJU4EWJHVCM5A4EVIQSXM7TCR5U26) 336 | 337 | 多模态行人检测的方法 338 | 339 | # 18. Multispectral Deep Neural Networks for Pedestrian Detection CCF C 340 | 341 | > ``` 342 | > @inproceedings{liu2016multispectral, 343 | > title={Multispectral deep neural networks for pedestrian detection}, 344 | > author={Liu, Jingjing and Zhang, Shaoting and Wang, Shu and Metaxas, Dimitris N}, 345 | > booktitle={27th British Machine Vision Conference, BMVC 2016}, 346 | > year={2016} 347 | > } 348 | > ``` 349 | 350 | ![image-20220526195617847](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526195617847.png?token=AJU4EWKYSFVHALALLYVUOTDCR5VSC) 351 | 352 | ![image-20220526195636994](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526195636994.png?token=AJU4EWO4NYXQKPLDH4VB73TCR5VTI) 353 | 354 | ![image-20220526195654140](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526195654140.png?token=AJU4EWJURGR4G6J7FR6A5IDCR5VUK) 355 | 356 | 也是多模态问题。提出了不同种融合方式的讨论 357 | 358 | # 19. Multi-interactive Dual-decoder for RGB-thermal Salient Object Detection CCF A 359 | 360 | > ``` 361 | > @article{tu2021multi, 362 | > title={Multi-Interactive dual-decoder for RGB-Thermal salient object detection}, 363 | > author={Tu, Zhengzheng and Li, Zhun and Li, Chenglong and Lang, Yang and Tang, Jin}, 364 | > journal={IEEE Transactions on Image Processing}, 365 | > volume={30}, 366 | > pages={5678--5691}, 367 | > year={2021}, 368 | > publisher={IEEE} 369 | > } 370 | > ``` 371 | 372 | ![Multi-interactive Dual-decoder for RGB-thermal](https://github.com/lengmo1996/PaperImageHost/blob/main/img/Multi-interactive%20Dual-decoder%20for%20RGB-thermal.jpg?raw=true) 373 | 374 | ![image-20220526211736614](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526211736614.png?token=AJU4EWOMIFBNNI2ZLNAQTNTCR57DA) 375 | 376 | RGB-T 的双模态显著图检测 377 | 378 | # 20. Motion and Region Aware Adversarial Learning for Fall Detection with Thermal Imaging CCF C 379 | 380 | > ``` 381 | > @inproceedings{mehta2021motion, 382 | > title={Motion and region aware adversarial learning for fall detection with thermal imaging}, 383 | > author={Mehta, Vineet and Dhall, Abhinav and Pal, Sujata and Khan, Shehroz S}, 384 | > booktitle={2020 25th International Conference on Pattern Recognition (ICPR)}, 385 | > pages={6321--6328}, 386 | > year={2021}, 387 | > organization={IEEE} 388 | > } 389 | > ``` 390 | 391 | ![image-20220526212008602](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526212008602.png?token=AJU4EWPTCLR7EBMILOLE3O3CR57MW) 392 | 393 | 这个主要是红外图像检测行人跌倒，跟我们主题不相关 394 | 395 | # 21. LLVIP: A Visible-infrared Paired Dataset for Low-light Vision (ICCV workshop) 396 | 397 | ![image-20220526212546641](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526212546641.png?token=AJU4EWIS4P6RUY5DHXGFAA3CR6ABW) 398 | 399 | 低照度行人检测数据集 400 | 401 | # 22. Linear Support Tensor Machine With LSK Channels: Pedestrian Detection in Thermal Infrared Images CCF A 402 | 403 | > ``` 404 | > @article{biswas2017linear, 405 | > title={Linear support tensor machine with LSK channels: Pedestrian detection in thermal infrared images}, 406 | > author={Biswas, Sujoy Kumar and Milanfar, Peyman}, 407 | > journal={IEEE transactions on image processing}, 408 | > volume={26}, 409 | > number={9}, 410 | > pages={4229--4242}, 411 | > year={2017}, 412 | > publisher={IEEE} 413 | > } 414 | > ``` 415 | 416 | ![image-20220526213806319](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526213806319.png?token=AJU4EWMLNFGC3EE257XL3K3CR6BP6) 417 | 418 | 17年的工作，是用传统方法来做的，在多个数据集上测试了性能，包括OSU，LSI和KAIST 419 | 420 | # 23. Illumination-aware faster R-CNN for robust multispectral pedestrian detection CCF B 421 | 422 | > ``` 423 | > @article{li2019illumination, 424 | > title={Illumination-aware faster R-CNN for robust multispectral pedestrian detection}, 425 | > author={Li, Chengyang and Song, Dan and Tong, Ruofeng and Tang, Min}, 426 | > journal={Pattern Recognition}, 427 | > volume={85}, 428 | > pages={161--171}, 429 | > year={2019}, 430 | > publisher={Elsevier} 431 | > } 432 | > ``` 433 | 434 | ![image-20220526215027554](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526215027554.png?token=AJU4EWPQF2PBBKZ4VKV6M5LCR6C6I) 435 | 436 | ![image-20220526215110959](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526215110959.png?token=AJU4EWM24Y233RBJ3SDZGPDCR6DA6) 437 | 438 | ![image-20220526215132849](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220526215132849.png?token=AJU4EWLJD4WNUBQ7DJNA5JDCR6DCI) 439 | 440 | 19年的工作，引用160+ 做的多模态行人检测，设计了两个光照感知网络，但是很像一篇TMM上的尺度感知行人检测工作 441 | 442 | # 24. Hierarchical Domain-Consistent Network For Cross-Domain Object Detection CCF C 443 | 444 | > ``` 445 | > @inproceedings{liu2021hierarchical, 446 | > title={Hierarchical Domain-Consistent Network For Cross-Domain Object Detection}, 447 | > author={Liu, Yuanyuan and Liu, Ziyang and Fang, Fang and Fu, Zhanghua and Chen, Zhanlong}, 448 | > booktitle={2021 IEEE International Conference on Image Processing (ICIP)}, 449 | > pages={474--478}, 450 | > year={2021}, 451 | > organization={IEEE} 452 | > } 453 | > ``` 454 | 455 | ![image-20220527163642318](C:\Users\lengmo\AppData\Roaming\Typora\typora-user-images\image-20220527163642318.png) 456 | 457 | 这个好像和红外相关性不强，不知道当时咋放进来了。做的是在不同域间转换的时候避免检测器性能损失。 458 | 459 | # 25. Fast Background Subtraction and Graph Cut for Thermal Pedestrian Detection CCF None 460 | 461 | > ``` 462 | > @inproceedings{oluyide2021fast, 463 | > title={Fast Background Subtraction and Graph Cut for Thermal Pedestrian Detection}, 464 | > author={Oluyide, Oluwakorede M and Tapamo, Jules-Raymond and Walingo, Tom}, 465 | > booktitle={Mexican Conference on Pattern Recognition}, 466 | > pages={219--228}, 467 | > year={2021}, 468 | > organization={Springer} 469 | > } 470 | > ``` 471 | 472 | 本文主要关注某一特定类型热成像仪（ ferroelectric thermal cameras ）。这种成像仪在成像时会产生信噪比低，在极冷或极热物体周围出现光晕或产生从白热到黑热的极性变化。 473 | 474 | # 26. Every feature counts: An improved one-stage detector in thermal imagery CCF None 475 | 476 | > ``` 477 | > @inproceedings{cao2019every, 478 | > title={Every feature counts: An improved one-stage detector in thermal imagery}, 479 | > author={Cao, Yu and Zhou, Tong and Zhu, Xinhua and Su, Yan}, 480 | > booktitle={2019 IEEE 5th International Conference on Computer and Communications (ICCC)}, 481 | > pages={1965--1969}, 482 | > year={2019}, 483 | > organization={IEEE} 484 | > } 485 | > ``` 486 | 487 | ![image-20220527171406210](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220527171406210.png?token=AJU4EWP7AARDC6YAVRCG7CLCSCLJY) 488 | 489 | ![image-20220527171445641](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220527171445641.png?token=AJU4EWITIVD2CCCXVBGYU4LCSCLMI) 490 | 491 | ![image-20220527171510087](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220527171510087.png?token=AJU4EWMAPM22HDBSBQIVKKTCSCLNY) 492 | 493 | 红外图像中不存在很多细节的视觉特性，例如颜色和文理，所以低级的特征和高级的特征对于提高检测器的性能是同样重要的。在RefineDet的基础上加了双路混合模块和逐通道增强模块。 494 | 495 | # 27. A Dual-branch Network for Infrared and Visible Image Fusion CCF C 496 | 497 | > ``` 498 | > @inproceedings{fu2021dual, 499 | > title={A dual-branch network for infrared and visible image fusion}, 500 | > author={Fu, Yu and Wu, Xiao-Jun}, 501 | > booktitle={2020 25th International Conference on Pattern Recognition (ICPR)}, 502 | > pages={10675--10680}, 503 | > year={2021}, 504 | > organization={IEEE} 505 | > } 506 | > ``` 507 | 508 | ![image-20220527193345761](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220527193345761.png?token=AJU4EWPY6VDPQIMO4R5UFZ3CSC3VQ) 509 | 510 | ![image-20220527193404025](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220527193404025.png?token=AJU4EWOVOYOQNXSA6RJ4JZTCSC3WU) 511 | 512 | 通过双流网络汇聚红外和可见光图像，这个图像聚合任务有点类似于重建？ 513 | 514 | # 28. Domain-adaptive pedestrian detection in thermal images CCF C 515 | 516 | > ``` 517 | > @inproceedings{guo2019domain, 518 | > title={Domain-adaptive pedestrian detection in thermal images}, 519 | > author={Guo, Tiantong and Huynh, Cong Phuoc and Solh, Mashhour}, 520 | > booktitle={2019 IEEE International Conference on Image Processing (ICIP)}, 521 | > pages={1660--1664}, 522 | > year={2019}, 523 | > organization={IEEE} 524 | > } 525 | > ``` 526 | 527 | > Under such a circumstance, 528 | > cues for human presence such as body shape, silhouette and 529 | > clothing textures are subject to the sensitivity of the imaging 530 | > sensors. 531 | 532 | ![image-20220528133134792](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528133134792.png?token=AJU4EWKM4DG6DZR2QXCMJY3CSGZ7K) 533 | 534 | ![image-20220528133156969](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528133156969.png?token=AJU4EWKOTONE2CKGUYXFJYLCSG2AW) 535 | 536 | 利用GAN网络生成大量虚拟红外图像和标注。将其与真实红外图像混合一起训练检测器。 537 | 538 | # 29. CNN-based thermal infrared person detection by domain adaptation CCF None 539 | 540 | > ``` 541 | > @inproceedings{herrmann2018cnn, 542 | > title={CNN-based thermal infrared person detection by domain adaptation}, 543 | > author={Herrmann, Christian and Ruf, Miriam and Beyerer, J{\"u}rgen}, 544 | > booktitle={Autonomous Systems: Sensors, Vehicles, Security, and the Internet of Everything}, 545 | > volume={10643}, 546 | > pages={1064308}, 547 | > year={2018}, 548 | > organization={International Society for Optics and Photonics} 549 | > } 550 | > ``` 551 | 552 | ![image-20220528133601064](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528133601064.png?token=AJU4EWMZIFGMOY5AGPGJD6LCSG2P6) 553 | 554 | 通过一些图像预处理方案使得红外图像尽可能地近似于可见光图像的灰度图，再丢进神经网络训练和微调。 555 | 556 | # 30. Borrow from Anywhere: Pseudo Multi-modal Object Detection in Thermal Imagery (CVPR Workshop) 557 | 558 | > ``` 559 | > @inproceedings{devaguptapu2019borrow, 560 | > title={Borrow from anywhere: Pseudo multi-modal object detection in thermal imagery}, 561 | > author={Devaguptapu, Chaitanya and Akolekar, Ninad and M Sharma, Manuj and N Balasubramanian, Vineeth}, 562 | > booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops}, 563 | > pages={0--0}, 564 | > year={2019} 565 | > } 566 | > ``` 567 | 568 | ![image-20220528134215784](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528134215784.png?token=AJU4EWJV52ARUJQQFWTRGJDCSG3HM) 569 | 570 | 提出了伪多模态检测的概念，将红外图像通过I2I框架转换为虚假的可见光图像，是用双流网络提取特征，合并后用于检测。 571 | 572 | # 31. Self-training Guided Adversarial Domain Adaptation For Thermal Imagery （CVPR 2021 Workshop） 573 | 574 | > ``` 575 | > @inproceedings{akkaya2021self, 576 | > title={Self-training guided adversarial domain adaptation for thermal imagery}, 577 | > author={Akkaya, Ibrahim Batuhan and Altinel, Fazil and Halici, Ugur}, 578 | > booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition}, 579 | > pages={4322--4331}, 580 | > year={2021} 581 | > } 582 | > ``` 583 | 584 | ![image-20220528135917798](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528135917798.png?token=AJU4EWLBTR5AOAT2H3K2OSTCSG5HG) 585 | 586 | ![image-20220528135949395](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528135949395.png?token=AJU4EWOP25LWKN67UFIFBE3CSG5JG) 587 | 588 | ![image-20220528140107704](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528140107704.png?token=AJU4EWIRYKICTWQD3IGFDWTCSG5OC) 589 | 590 | 无监督域适应能够学到一个不需要图像对就可以把两个域映射到同一个特征空间中的模型。通过学习域不变特征标识，对抗域适应假设一个在源域上训练的分类器可以同样在目标域上有效。 591 | 592 | ![image-20220528140414877](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528140414877.png?token=AJU4EWJXSLJPMOOURRSWCR3CSG5ZY) 593 | 594 | # 32. Bottom-up and layerwise domain adaptation for pedestrian detection in thermal images CCF B 595 | 596 | > ``` 597 | > @article{kieu2021bottom, 598 | > title={Bottom-up and layerwise domain adaptation for pedestrian detection in thermal images}, 599 | > author={Kieu, My and Bagdanov, Andrew D and Bertini, Marco}, 600 | > journal={ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM)}, 601 | > volume={17}, 602 | > number={1}, 603 | > pages={1--19}, 604 | > year={2021}, 605 | > publisher={ACM New York, NY} 606 | > } 607 | > ``` 608 | 609 | ![image-20220528141103987](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528141103987.png?token=AJU4EWPBVSRXNKP4JBEHFI3CSG6TK) 610 | 611 | ![image-20220528141120277](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528141120277.png?token=AJU4EWPX2SBK3XWU4FXO2VLCSG6UM) 612 | 613 | 提出了自顶向下的域适应方法，将在可见光域训练好的模型适应到红外域中。剔除了自底向上的域适应方法，训练一个适配器段，初始化红外检测器的层以适应新的输入的分布，再将适配器重新链接到原始的可见光检测器，以进行自上而下的损失。 614 | 615 | ![image-20220528142222642](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528142222642.png?token=AJU4EWKZ4YOZXBN4YZ44SOTCSG752) 616 | 617 | # 33. Domain Adaptation for Privacy-preserving Pedestrian Detection in Thermal Imagery CCF None 618 | 619 | > ``` 620 | > @inproceedings{kieu2019domain, 621 | > title={Domain adaptation for privacy-preserving pedestrian detection in thermal imagery}, 622 | > author={Kieu, My and Bagdanov, Andrew D and Bertini, Marco and Bimbo, Alberto Del}, 623 | > booktitle={International Conference on Image Analysis and Processing}, 624 | > pages={203--213}, 625 | > year={2019}, 626 | > organization={Springer} 627 | > } 628 | > ``` 629 | 630 | ![image-20220528142519835](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528142519835.png?token=AJU4EWKZFND2VNHYNX7SKADCSHAI2) 631 | 632 | ![image-20220528142640599](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/image-20220528142640599.png?token=AJU4EWKH2ZEM6JIUS72LC5TCSHAN4) 633 | 634 | 研究了两种简单域适应策略在红外图像检测上的潜力 635 | 636 | # 34. Exploring thermal images for object detection in underexposure regions for autonomous driving 637 | 638 | > ``` 639 | > @article{munir2022exploring, 640 | > title={Exploring thermal images for object detection in underexposure regions for autonomous driving}, 641 | > author={Munir, Farzeen and Azam, Shoaib and Rafique, Muhammd Aasim and Sheri, Ahmad Muqeem and Jeon, Moongu and Pedrycz, Witold}, 642 | > journal={Applied Soft Computing}, 643 | > volume={121}, 644 | > pages={108793}, 645 | > year={2022}, 646 | > publisher={Elsevier} 647 | > } 648 | > ``` 649 | 650 | Another version of # 3 651 | 652 | # 35. Salient Features for Moving Object Detection in Adverse Weather Conditions During Night Time CCF B 653 | > ``` 654 | > @article{singha2019salient, 655 | > title={Salient features for moving object detection in adverse weather conditions during night time}, 656 | > author={Singha, Anu and Bhowmik, Mrinal Kanti}, 657 | > journal={IEEE Transactions on circuits and systems for video technology}, 658 | > volume={30}, 659 | > number={10}, 660 | > pages={3317--3331}, 661 | > year={2019}, 662 | > publisher={IEEE} 663 | > } 664 | > ``` 665 | 666 | ![image-20220606101233692](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061012060.png) 667 | 668 | 提出了一个新的数据集，夜间视频数据集，命名为TU-VDN，用于热红外图像中的移动物体检测。提出了一个基于视频显著性背景分割技术，同时使用红外帧中的空域特征和热强度。 669 | 670 | # 36. A paced multi-stage block-wise approach for object detection in thermal images CCF C 671 | > ``` 672 | > @article{kera2022paced, 673 | > title={A paced multi-stage block-wise approach for object detection in thermal images}, 674 | > author={Kera, Shreyas Bhat and Tadepalli, Anand and Ranjani, J Jennifer}, 675 | > journal={The Visual Computer}, 676 | > pages={1--17}, 677 | > year={2022}, 678 | > publisher={Springer} 679 | > } 680 | > ``` 681 | 682 | ![image-20220606101848368](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061019028.png) 683 | 684 | ![image-20220606101905522](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061019634.png) 685 | 686 | ![image-20220606101929263](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061019324.png) 687 | 688 | # 37. Robust Thermal Infrared Pedestrian Detection By Associating Visible Pedestrian Knowledge CCF C 689 | 690 | > ``` 691 | > @inproceedings{park2022robust, 692 | > title={Robust Thermal Infrared Pedestrian Detection By Associating Visible Pedestrian Knowledge}, 693 | > author={Park, Sungjune and Choi, Dae Hwi and Kim, Jung Uk and Ro, Yong Man}, 694 | > booktitle={ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, 695 | > pages={4468--4472}, 696 | > year={2022}, 697 | > organization={IEEE} 698 | > } 699 | > ``` 700 | 701 | ![image-20220606102323808](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061023016.png) 702 | 703 | ![image-20220606102337643](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061023720.png) 704 | 705 | # 38. Thermal Object Detection in Difficult Weather Conditions Using YOLO 706 | > ``` 707 | > @article{krivsto2020thermal, 708 | > title={Thermal object detection in difficult weather conditions using YOLO}, 709 | > author={Kri{\v{s}}to, Mate and Ivasic-Kos, Marina and Pobar, Miran}, 710 | > journal={IEEE access}, 711 | > volume={8}, 712 | > pages={125459--125476}, 713 | > year={2020}, 714 | > publisher={IEEE} 715 | > } 716 | > ``` 717 | 718 | ![image-20220606102540719](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061025840.png) 719 | 720 | ![image-20220606102606986](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061026073.png) 721 | 722 | # 39. Infrared Image Saliency Detection based on Human Vision and Information Theory CCF None 723 | > ``` 724 | > @inproceedings{qin2016infrared, 725 | > title={Infrared image saliency detection based on human vision and information theory}, 726 | > author={Qin, Shiyu and Wang, Lan and Cheng, Hua and Feng, Qi and Zhang, Minwen and Gao, Chenqiang}, 727 | > booktitle={2016 9th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)}, 728 | > pages={484--488}, 729 | > year={2016}, 730 | > organization={IEEE} 731 | > } 732 | > ``` 733 | 734 | 看起来比较奇怪，虽然公式很多，但全程没有图片和图表 735 | 736 | # 40. TIRNet: Object detection in thermal infrared images for autonomous driving CCF C 737 | > ``` 738 | > @article{dai2021tirnet, 739 | > title={TIRNet: Object detection in thermal infrared images for autonomous driving}, 740 | > author={Dai, Xuerui and Yuan, Xue and Wei, Xueye}, 741 | > journal={Applied Intelligence}, 742 | > volume={51}, 743 | > number={3}, 744 | > pages={1244--1261}, 745 | > year={2021}, 746 | > publisher={Springer} 747 | > } 748 | > ``` 749 | 750 | ![image-20220606102850616](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061028728.png) 751 | 752 | # 41. Pedestrian Detection at Night in Infrared Images Using an Attention-Guided Encoder-Decoder Convolutional Neural Network CCF None 753 | > ``` 754 | > @article{chen2020pedestrian, 755 | > title={Pedestrian detection at night in infrared images using an attention-guided encoder-decoder convolutional neural network}, 756 | > author={Chen, Yunfan and Shin, Hyunchul}, 757 | > journal={Applied Sciences}, 758 | > volume={10}, 759 | > number={3}, 760 | > pages={809}, 761 | > year={2020}, 762 | > publisher={Multidisciplinary Digital Publishing Institute} 763 | > } 764 | > ``` 765 | 766 | ![image-20220606103012895](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061030012.png) 767 | 768 | ![image-20220606103031298](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061030392.png) 769 | 770 | # 42. Borrow from Source Models: Efficient Infrared Object Detection with Limited Examples CCF None 771 | > ``` 772 | > @article{chen2022borrow, 773 | > title={Borrow from Source Models: Efficient Infrared Object Detection with Limited Examples}, 774 | > author={Chen, Ruimin and Liu, Shijian and Mu, Jing and Miao, Zhuang and Li, Fanming}, 775 | > journal={Applied Sciences}, 776 | > volume={12}, 777 | > number={4}, 778 | > pages={1896}, 779 | > year={2022}, 780 | > publisher={MDPI} 781 | > } 782 | > ``` 783 | 784 | ![image-20220606104856969](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061048082.png) 785 | 786 | # 43. Feature distribution alignments for object detection in the thermal domain CCF C 787 | 788 | > ``` 789 | > @article{marnissi2022feature, 790 | > title={Feature distribution alignments for object detection in the thermal domain}, 791 | > author={Marnissi, Mohamed Amine and Fradi, Hajer and Sahbani, Anis and Essoukri Ben Amara, Najoua}, 792 | > journal={The Visual Computer}, 793 | > pages={1--13}, 794 | > year={2022}, 795 | > publisher={Springer} 796 | > } 797 | > ``` 798 | 799 | ![image-20220606104959340](https://raw.githubusercontent.com/lengmo1996/PaperImageHost/main/img/202206061049513.png) 800 | 801 | # 44. 802 | --------------------------------------------------------------------------------