└── README.md /README.md: -------------------------------------------------------------------------------- 1 | Noisy Labels in Computer Vision 2 | ========= 3 | 4 | A curated list of papers that study learning with noisy labels. 5 | 6 | --- 7 | 8 | - [Image Classification](#image-classification) 9 | - [GitHub Repository](#github-repository) 10 | - [Survey](#survey) 11 | - [Distinguished Researchers and Team](#distinguished-researchers-and-team) 12 | - [Object Detection](#object-detection) 13 | - [Segmentation](#segmentation) 14 | - [Object Counting](#object-counting) 15 | 16 | 17 | --- 18 | 19 | Image Classification 20 | ==================== 21 | 22 | GitHub Repository 23 | --- 24 | * [[Awesome-Learning-with-Label-Noise]](https://github.com/subeeshvasu/Awesome-Learning-with-Label-Noise) ![GitHub Repo stars](https://img.shields.io/github/stars/subeeshvasu/Awesome-Learning-with-Label-Noise?style=social) 25 | 26 | Survey 27 | --- 28 | 29 | * [**2014 TNNLS**] Classification in the Presence of Label Noise: A Survey [[paper]](https://ieeexplore.ieee.org/document/6685834%5C%22) 30 | * [**2019 KBS**] Image Classification with Deep Learning in the Presence of Noisy Labels: A Survey [[paper]](https://arxiv.org/abs/1912.05170) 31 | * [**2020 MIA**] Deep learning with noisy labels: exploring techniques and remedies in medical image analysis [[paper]](https://www.sciencedirect.com/science/article/pii/S1361841520301237) 32 | * [**2020 ArXiv**] A Survey of Label-noise Representation Learning: Past, Present and Future [[paper]](https://arxiv.org/abs/2011.04406) [[code]](https://github.com/bhanML/label-noise-papers) ![GitHub Repo stars](https://img.shields.io/github/stars/bhanML/label-noise-papers?style=social) 33 | * [**2022 TNNLS**] Learning from Noisy Labels with Deep Neural Networks: A Survey [[paper]](https://arxiv.org/abs/2007.08199) [[code]](https://github.com/songhwanjun/Awesome-Noisy-Labels) ![GitHub Repo stars](https://img.shields.io/github/stars/songhwanjun/Awesome-Noisy-Labels?style=social) 34 | 35 | Distinguished Researchers and Team 36 | --- 37 | * [Tongliang Liu](https://tongliang-liu.github.io/), The University of Sydney 38 | * [Bo Han](https://bhanml.github.io/), Hong Kong Baptist University 39 | * [Yang Liu](http://www.yliuu.com/), UC Santa Cruz 40 | * [RIKEN-AIP](https://aip.riken.jp/labs/generic_tech/imperfect_inf_learn/), Japan 41 | 42 | 43 | Object Detection 44 | ================ 45 | 46 | 2024 47 | --- 48 | * [**ArXiv**] **DN-TOD**: Haoran Zhua, Chang Xua, Wen Yanga, Ruixiang Zhanga, Yan Zhanga, Gui-Song Xia. 49 | "Robust Tiny Object Detection in Aerial Images amidst Label Noise." 50 | [[paper]](https://arxiv.org/pdf/2401.08056.pdf) 51 | 52 | 53 | 2023 54 | ---- 55 | * [**ICCV 2023**] **SSD-Det**: Di Wu, Pengfei Chen, Xuehui Yu, Guorong Li, Zhenjun Han, Jianbin Jiao. 56 | "Spatial Self-Distillation for Object Detection with Inaccurate Bounding Boxes." 57 | [[paper]](https://arxiv.org/pdf/2307.12101v1.pdf) 58 | [[code]](https://github.com/ucas-vg/PointTinyBenchmark/tree/SSD-Det) 59 | 60 | * [**ArXiv 2023**] Donghao Zhou, Jialin Li, Jinpeng Li, Jiancheng Huang, Qiang Nie, Yong Liu, Bin-Bin Gao, Qiong Wang, Pheng-Ann Heng, Guangyong Chen. 61 | "Distribution-Aware Calibration for Object Detection with Noisy Bounding Boxes." [[paper]](https://arxiv.org/pdf/2308.12017v1.pdf) 62 | 63 | * [**ArXiv 2023**] Marius Schubert, Tobias Riedlinger, Karsten Kahl, Daniel Kröll, Sebastian Schoenen, Siniša Šegvic, Matthias Rottmann. 64 | "Identifying Label Errors in Object Detection Datasets by Loss Inspection." [[paper]](https://arxiv.org/pdf/2308.12017v1.pdf](https://arxiv.org/pdf/2303.06999.pdf)) 65 | 66 | * [**ArXiv 2023**] **UNA**: Kwangrok Ryoo, Yeonsik Jo, Seungjun Lee, Mira Kim, Ahra Jo, Seung Hwan Kim, Seungryong Kim, Soonyoung Lee. 67 | "Universal Noise Annotation: Unveiling the Impact of Noisy Annotation on Object Detection." 68 | [[paper]](https://arxiv.org/pdf/2312.13822.pdf) 69 | [[code]](https://github.com/Ryoo72/UNA) 70 | ![GitHub Repo stars](https://img.shields.io/github/stars/Ryoo72/UNA?style=social) 71 | 72 | 2022 73 | ---- 74 | 75 | * [**CVPR 2022**] **NLTE**: Xinyu Liu, Wuyang Li, Qiushi Yang, Baopu Li, Yixuan Yuan. 76 | "Towards Robust Adaptive Object Detection under Noisy Annotations." 77 | [[paper]](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_Towards_Robust_Adaptive_Object_Detection_Under_Noisy_Annotations_CVPR_2022_paper.pdf) 78 | [[code]](https://github.com/CityU-AIM-Group/NLTE) 79 | ![GitHub Repo stars](https://img.shields.io/github/stars/CityU-AIM-Group/NLTE?style=social) 80 | 81 | * [**ECCV 2022**] **OA-MIL**: Chengxin Liu, Kewei Wang, Hao Lu, Zhiguo Cao, Ziming Zhang. 82 | "Robust Object Detection With Inaccurate Bounding Boxes." 83 | [[paper]](https://arxiv.org/pdf/2207.09697.pdf) 84 | [[code]](https://github.com/cxliu0/OA-MIL) 85 | ![GitHub Repo stars](https://img.shields.io/github/stars/cxliu0/OA-MIL?style=social) 86 | 87 | * [**ECCV 2022**] **W2N**: Zitong Huang, Yiping Bao, Bowen Dong, Erjin Zhou, Wangmeng Zuo. 88 | "W2N: Switching From Weak Supervision to Noisy Supervision for Object Detection." 89 | [[paper]](https://arxiv.org/pdf/2207.12104.pdf) 90 | [[code]](https://github.com/1170300714/w2n_wsod) 91 | ![GitHub Repo stars](https://img.shields.io/github/stars/1170300714/w2n_wsod?style=social) 92 | 93 | * [**Remote Sensing 2022**] Maximilian Bernhard, Matthias Schubert. 94 | "Correcting Imprecise Object Locations for Training Object Detectors in Remote Sensing Applications." [[paper]](https://www.mdpi.com/2072-4292/13/24/4962) 95 | 96 | * [**TIP 2022**] Shaoru Wang, Jin Gao, Bing Li, Weiming Hu. 97 | "Narrowing the Gap: Improved Detector Training with Noisy Location Annotations." [[paper]](https://arxiv.org/pdf/2206.05708.pdf) 98 | 99 | * [**ArXiv 2022**] Krystian Chachuła, Adam Popowicz, Jakub Łyskawa, Bartłomiej Olber, Piotr Fr ̨ atczak, Krystian Radlak. 100 | "Combating noisy labels in object detection datasets." [[paper]](https://arxiv.org/pdf/2211.13993.pdf) 101 | 102 | * [**ACM GIS 2022**] Maximilian Bernhard, Matthias Schubert. 103 | "Robust object detection in remote sensing imagery with noisy and sparse geo-annotations." 104 | [[paper]](https://arxiv.org/pdf/2210.12989.pdf) 105 | [[code]](https://github.com/mxbh/robust_object_detection) 106 | ![GitHub Repo stars](https://img.shields.io/github/stars/mxbh/robust_object_detection?style=social) 107 | 108 | 2021 109 | ---- 110 | 111 | * [**TIP 2021**] **MRNet**: Youjiang Xu, Linchao Zhu, YiYang, Fei Wu. 112 | "Training Robust Object Detectors From Noisy Category Labels and Imprecise Bounding Boxes." 113 | [[paper]](https://ieeexplore.ieee.org/document/9457066) 114 | 115 | * [**BMVC 2021**] Jiafeng Mao, Qing Yu, Yoko Yamakata, Kiyoharu Aizawa. 116 | "Noisy Annotation Refinement for Object Detection" [[paper]](https://www.bmvc2021-virtualconference.com/assets/papers/0778.pdf) 117 | 118 | * [**IEICE TIS 2021**] Jiafeng Mao, Qing Yu, Kiyoharu Aizawa. 119 | "Noisy Localization Annotation Refinement for Object Detection." 120 | [[paper]](https://www.jstage.jst.go.jp/article/transinf/E104.D/9/E104.D_2021EDP7026/_pdf) 121 | 122 | 123 | 2020 124 | ---- 125 | 126 | * [**CVPR 2020**] Yunhang Shen, Rongrong Ji, Zhiwei Chen, Xiaopeng Hong, Feng Zheng, Jianzhuang Liu, Mingliang Xu, Qi Tian. 127 | "Noise-Aware Fully Webly Supervised Object Detection." 128 | [[paper]](https://openaccess.thecvf.com/content_CVPR_2020/papers/Shen_Noise-Aware_Fully_Webly_Supervised_Object_Detection_CVPR_2020_paper.pdf) 129 | [[code]](https://github.com/shenyunhang/NA-fWebSOD) 130 | ![GitHub Repo stars](https://img.shields.io/github/stars/shenyunhang/NA-fWebSOD?style=social) 131 | 132 | * [**CVPR 2020**] Hengduo Li, Zuxuan Wu, Chen Zhu, Caiming Xiong, Richard Socher, Larry S. Davis. 133 | "Learning From Noisy Anchors for One-Stage Object Detection." 134 | [[paper]](https://openaccess.thecvf.com/content_CVPR_2020/papers/Li_Learning_From_Noisy_Anchors_for_One-Stage_Object_Detection_CVPR_2020_paper.pdf) 135 | 136 | * [**CVPRW 2020**] Aybora Koksal, Kutalmis Gokalp Ince, A. Aydin Alatan. 137 | "Effect of Annotation Errors on Drone Detection with YOLOv3." 138 | [[paper]](https://openaccess.thecvf.com/content_CVPRW_2020/papers/w69/Koksal_Effect_of_Annotation_Errors_on_Drone_Detection_With_YOLOv3_CVPRW_2020_paper.pdf) 139 | 140 | * [**ICIP 2020**] Jiafeng Mao, Qing Yu, Kiyoharu Aizawa. 141 | "Noisy Localization Annotation Refinement For Object Detection." [[paper]](https://ieeexplore.ieee.org/document/9190728) 142 | 143 | * [**ArXiv 2020**] Junnan Li, Caiming Xiong, Richard Socher, Steven Hoi. 144 | "Towards Noise-resistant Object Detection with Noisy Annotations." [[paper]](https://arxiv.org/pdf/2003.01285.pdf) 145 | 146 | 147 | 2019 148 | ---- 149 | 150 | * [**ICCV 2019**] **NOTE-RCNN**: Jiyang Gao, Jiang Wang, Shengyang Dai, Li-Jia Li, Ram Nevatia. 151 | "NOTE-RCNN: NOise Tolerant Ensemble RCNN for Semi-Supervised Object Detection." 152 | [[paper]](https://openaccess.thecvf.com/content_ICCV_2019/papers/Gao_NOTE-RCNN_NOise_Tolerant_Ensemble_RCNN_for_Semi-Supervised_Object_Detection_ICCV_2019_paper.pdf) 153 | 154 | * [**AAAI 2019**] **SD-LocNet**: Xiaopeng Zhang, Yang Yang, Jiashi Feng. 155 | "Learning to Localize Objects with Noisy Labeled Instances." 156 | [[paper]](https://ojs.aaai.org/index.php/AAAI/article/view/4957) 157 | 158 | * [**IV 2019**] Simon Chadwick, Paul Newman. 159 | "Training object detectors with noisy data." [[paper]](https://arxiv.org/pdf/1905.07202.pdf) 160 | 161 | Segmentation 162 | ============ 163 | 164 | 2023 165 | ---- 166 | * [**ICLR 2023**] Jiachen Yao, Yikai Zhang, Songzhu Zheng, Mayank Goswami, Prateek Prasanna, Chao Chen. 167 | "Learning to Segment From Noisy Annotations: A Spatial Correction Approach." 168 | [[paper]](https://openreview.net/pdf?id=Qc_OopMEBnC) 169 | [[code]](https://github.com/michaelofsbu/SpatialCorrection) 170 | ![GitHub Repo stars](https://img.shields.io/github/stars/michaelofsbu/SpatialCorrection?style=social) 171 | 172 | * [**ArXiv 2023**] Zicheng Wang, Zhen Zhao, Erjian Guo, Luping Zhou. 173 | "Clean Label Disentangling for Medical Image Segmentation with Noisy Labels." 174 | [[paper]](https://arxiv.org/pdf/2311.16580.pdf) 175 | [[code]](https://github.com/xiaoyao3302/2BDenoise) 176 | 177 | 2022 178 | ---- 179 | * [**CVPR 2022 oral**] Sheng Liu, Kangning Liu, Weicheng Zhu, Yiqiu Shen, Carlos Fernandez-Granda. 180 | "Adaptive Early-Learning Correction for Segmentation from Noisy Annotations." 181 | [[paper]](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_Adaptive_Early-Learning_Correction_for_Segmentation_From_Noisy_Annotations_CVPR_2022_paper.pdf) 182 | [[code]](https://github.com/Kangningthu/ADELE) 183 | ![GitHub Repo stars](https://img.shields.io/github/stars/Kangningthu/ADELE?style=social) 184 | 185 | * [**CVPR 2022**] **SimT**: Xiaoqing Guo, Jie Liu, Tongliang Liu, Yixuan Yuan. 186 | "SimT: Handling Open-set Noise for Domain Adaptive Semantic Segmentation." 187 | [[paper]](https://openaccess.thecvf.com/content/CVPR2022/papers/Guo_SimT_Handling_Open-Set_Noise_for_Domain_Adaptive_Semantic_Segmentation_CVPR_2022_paper.pdf) 188 | [[code]](https://github.com/CityU-AIM-Group/SimT) 189 | ![GitHub Repo stars](https://img.shields.io/github/stars/CityU-AIM-Group/SimT?style=social) 190 | * (TPAMI version) Handling Open-set Noise and Novel Target Recognition in Domain Adaptive Semantic Segmentation. [[paper]](https://ieeexplore.ieee.org/abstract/document/10048580) 191 | 192 | * [**AAAI 2022**] Yaoru Luo, Guole Liu, Yuanhao Guo, Ge Yang. 193 | "Deep Neural Networks Learn Meta-Structures from Noisy Labels in Semantic Segmentation." 194 | [[paper]](https://www.aaai.org/AAAI22Papers/AAAI-12729.LuoY.pdf) 195 | [[code]](https://github.com/YaoruLuo/Meta-Structures-for-DNN) 196 | ![GitHub Repo stars](https://img.shields.io/github/stars/YaoruLuo/Meta-Structures-for-DNN?style=social) 197 | 198 | 2021 199 | ---- 200 | * [**CVPR 2021**] Youngmin Oh, Beomjun Kim, Bumsub Ham. 201 | "Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentation." 202 | [[paper]](https://openaccess.thecvf.com/content/CVPR2021/papers/Oh_Background-Aware_Pooling_and_Noise-Aware_Loss_for_Weakly-Supervised_Semantic_Segmentation_CVPR_2021_paper.pdf) 203 | [[code]](https://github.com/cvlab-yonsei/BANA) 204 | ![GitHub Repo stars](https://img.shields.io/github/stars/cvlab-yonsei/BANA?style=social) 205 | 206 | * [**ICCV 2021 oral**] Shuquan Ye, Dongdong Chen, Songfang Han, Jing Liao. 207 | "Learning with Noisy Labels for Robust Point Cloud Segmentation." 208 | [[paper]](https://openaccess.thecvf.com/content/ICCV2021/papers/Ye_Learning_With_Noisy_Labels_for_Robust_Point_Cloud_Segmentation_ICCV_2021_paper.pdf) 209 | [[code]](https://github.com/pleaseconnectwifi/PNAL) 210 | ![GitHub Repo stars](https://img.shields.io/github/stars/pleaseconnectwifi/PNAL?style=social) 211 | * (TPAMI version) Robust Point Cloud Segmentation with Noisy Annotations. [[paper]](https://ieeexplore.ieee.org/document/9966842/) 212 | 213 | * [**ICCV 2021**] Yuxi Wang, Junran Peng, Zhaoxiang Zhang. 214 | "Uncertainty-aware Pseudo Label Refinery for Domain Adaptive Semantic Segmentation." 215 | [[paper]](https://openaccess.thecvf.com/content/ICCV2021/papers/Wang_Uncertainty-Aware_Pseudo_Label_Refinery_for_Domain_Adaptive_Semantic_Segmentation_ICCV_2021_paper.pdf) 216 | 217 | * [**IJCV 2021**] Zhedong Zheng, Yi Yang. 218 | "Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation." 219 | [[paper]](https://link.springer.com/article/10.1007/s11263-020-01395-y) 220 | [[code]](https://github.com/layumi/Seg-Uncertainty) 221 | ![GitHub Repo stars](https://img.shields.io/github/stars/layumi/Seg-Uncertainty?style=social) 222 | * [**IJCAI 2020 Conference Version**] Unsupervised Scene Adaptation with Memory Regularization in vivo [[paper]](https://arxiv.org/pdf/1912.11164.pdf) 223 | 224 | 2020 225 | ---- 226 | * [**ECCV 2020**] Longrong Yang, Fanman Meng, Hongliang Li, Qingbo Wu, Qishang Cheng. 227 | "Learning with Noisy Class Labels for Instance Segmentation." 228 | [[paper]](https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123590035.pdf) 229 | [[code]](https://github.com/longrongyang/Learning-with-Noisy-Class-Labels-for-Instance-Segmentation) 230 | ![GitHub Repo stars](https://img.shields.io/github/stars/longrongyang/LNCIS?style=social) 231 | 232 | * [**NeurIPS 2020**] Le Zhang, Ryutaro Tanno, Mou-Cheng Xu, Chen Jin, Joseph Jacob, Olga Ciccarelli, Frederik Barkhof, Daniel C. Alexander. 233 | "Disentangling Human Error from the Ground Truth in Segmentation of Medical Images." 234 | [[paper]](https://proceedings.neurips.cc/paper/2020/file/b5d17ed2b502da15aa727af0d51508d6-Paper.pdf) 235 | [[code]](https://github.com/moucheng2017/Learn_Noisy_Labels_Medical_Images) 236 | ![GitHub Repo stars](https://img.shields.io/github/stars/moucheng2017/Learn_Noisy_Labels_Medical_Images?style=social) 237 | 238 | * [**MICCAI 2020**] Minqing Zhang, Jiantao Gao, Zhen Lyu, Weibing Zhao, Qin Wang, Weizhen Ding, Sheng Wang, Zhen Li, Shuguang Cui. 239 | "Characterizing Label Errors: Confident Learning for Noisy-labeled Image Segmentation." 240 | [[paper]](https://link.springer.com/chapter/10.1007/978-3-030-59710-8_70) 241 | [[code]](https://github.com/502463708/Confident_Learning_for_Noisy-labeled_Medical_Image_Segmentation) 242 | ![GitHub Repo stars](https://img.shields.io/github/stars/502463708/Confident_Learning_for_Noisy-labeled_Medical_Image_Segmentation?style=social) 243 | 244 | 2019 245 | --- 246 | 247 | * [**CVPR 2019**] Yi Zhu, Karan Sapra, Fitsum A. Reda, Kevin J. Shih, Shawn Newsam, Andrew Tao, Bryan Catanzaro. 248 | "Improving Semantic Segmentation via Video Propagation and Label Relaxation." 249 | [[paper]](https://openaccess.thecvf.com/content_CVPR_2019/papers/Zhu_Improving_Semantic_Segmentation_via_Video_Propagation_and_Label_Relaxation_CVPR_2019_paper.pdf) 250 | [[code]](https://github.com/NVIDIA/semantic-segmentation/tree/sdcnet) 251 | 252 | 253 | Object Counting 254 | =============== 255 | 256 | 2023 257 | --- 258 | 259 | * [**TPAMI 2023**] Jia Wan, Qiangqiang Wu, Antoni B. Chan. 260 | "Modeling Noisy Annotations for Point-wise Supervision." 261 | [[paper]](https://ieeexplore.ieee.org/abstract/document/10197253/authors#authors) 262 | 263 | * [**ArXiv 2023**] **SACC-Net**: Yi-Kuan Hsieh, Jun-Wei Hsieh, Xin li, Ming-Ching Chang, Yu-Chee Tseng. 264 | "Scale-Aware Crowd Count Network with Annotation Error Correction." 265 | [[paper]](https://arxiv.org/pdf/2312.16771.pdf) 266 | 267 | * [**ArXiv 2023**] Yuda Zou, Xin Xiao, Peilin Zhou, Zhichao Sun, Bo Du, Yongchao Xu. 268 | "Noised Autoencoders for Point Annotation Restoration in Object Counting." 269 | [[paper]](https://arxiv.org/pdf/2312.07190.pdf) 270 | 271 | * [**ArXiv 2023**] Yuehai Chen, Jing Yang, Badong Chen, Shaoyi Du, Gang Hua. 272 | "Point Annotation Probability Map: Towards Dense Object Counting by Tolerating Annotation Noise." 273 | [[paper]](https://arxiv.org/pdf/2308.00530.pdf) 274 | 275 | 2022 and before 276 | --- 277 | * [**CVPR 2022**] Zhi-Qi Cheng, Qi Dai, Hong Li, Jingkuan Song, Xiao Wu, Alexander G. Hauptmann. 278 | "Rethinking Spatial Invariance of Convolutional Networks for Object Counting." 279 | [[paper]](https://openaccess.thecvf.com/content/CVPR2022/papers/Cheng_Rethinking_Spatial_Invariance_of_Convolutional_Networks_for_Object_Counting_CVPR_2022_paper.pdf) 280 | [[code]](https://github.com/zhiqic/Rethinking-Counting) 281 | ![GitHub Repo stars](https://img.shields.io/github/stars/zhiqic/Rethinking-Counting?style=social) 282 | 283 | * [**NeurIPS 2020**] Jia Wan, Antoni B. Chan. 284 | "Modeling Noisy Annotations for Crowd Counting." 285 | [[paper]](https://proceedings.neurips.cc/paper/2020/file/22bb543b251c39ccdad8063d486987bb-Paper.pdf) 286 | [[code]](https://github.com/jia-wan/NoisyCC-pytorch) 287 | ![GitHub Repo stars](https://img.shields.io/github/stars/jia-wan/NoisyCC-pytorch?style=social) 288 | 289 | --------------------------------------------------------------------------------