mlsanityData to 7 Oct 2026

DetCo: Unsupervised Contrastive Learning for Object Detection

arXiv 2102.04803 ↗First posted 9 Feb 2021cs.CVcited by 205

Not on the map. Why?

Abstract (arXiv)

Unsupervised contrastive learning achieves great success in learning image representations with CNN. Unlike most recent methods that focused on improving accuracy of image classification, we present a novel contrastive learning approach, named DetCo, which fully explores the contrasts between global image and local image patches to learn discriminative representations for object detection. DetCo has several appealing benefits. (1) It is carefully designed by investigating the weaknesses of current self-supervised methods, which discard important representations for object detection. (2) DetCo builds hierarchical intermediate contrastive losses between global image and local patches to improve object detection, while maintaining global representations for image recognition. Theoretical analysis shows that the local patches actually remove the contextual information of an image, improving the lower bound of mutual information for better contrastive learning. (3) Extensive experiments on PASCAL VOC, COCO and Cityscapes demonstrate that DetCo not only outperforms state-of-the-art methods on object detection, but also on segmentation, pose estimation, and 3D shape prediction, while it is still competitive on image classification. For example, on PASCAL VOC, DetCo-100ep achieves 57.4 mAP, which is on par with the result of MoCov2-800ep. Moreover, DetCo consistently outperforms supervised method by 1.6/1.2/1.0 AP on Mask RCNN-C4/FPN/RetinaNet with 1x schedule. Code will be released at hrefhttps://github.com/xieenze/DetCocolorbluett github.com/xieenze/DetCo.

Key facts. Cited by 205 papers in the map (data through 7 Oct 2026). Builds on 39 of its 46 references that are in the map. First posted on arXiv 9 Feb 2021.

Builds on

39 of 46 references.

References · 39

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  4. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  5. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  6. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  7. 2017Mask R-CNNcited by 7,259
  8. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  9. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  10. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  11. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  12. 2013Auto-Encoding Variational Bayescited by 3,980
27 more
  1. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,845
  2. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  3. 2016Pyramid Scene Parsing Networkcited by 2,472
  4. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,282
  5. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,261
  6. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
  7. 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,838
  8. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,586
  9. 2016Context Encoders: Feature Learning by Inpaintingcited by 1,424
  10. 2016Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzlescited by 1,379
  11. 2016Colorful Image Colorizationcited by 1,113
  12. 2015Unsupervised Visual Representation Learning by Context Predictioncited by 1,062
  13. 2019Self-Supervised Learning of Pretext-Invariant Representationscited by 806
  14. 2017Focal Loss for Dense Object Detectioncited by 749
  15. 2018DensePose: Dense Human Pose Estimation In The Wildcited by 573
  16. 2020Dense Contrastive Learning for Self-Supervised Visual Pre-Trainingcited by 512
  17. 2020Sparse R-CNN: End-to-End Object Detection with Learnable Proposalscited by 494
  18. 2018Rethinking ImageNet Pre-trainingcited by 376
  19. 2016Adversarial Feature Learningcited by 350
  20. 2014Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networkscited by 299
  21. 2016Learning Representations for Automatic Colorizationcited by 251
  22. 2019Large Scale Adversarial Representation Learningcited by 238
  23. 2016Adversarially Learned Inferencecited by 216
  24. 2019Local Aggregation for Unsupervised Learning of Visual Embeddingscited by 204
  25. 2017Multi-task Self-Supervised Visual Learningcited by 200
  26. 2017MegDet: A Large Mini-Batch Object Detectorcited by 66
  27. 2014Stochastic Backpropagation and Approximate Inference in Deep Generative Modelscited by 30

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 32; 2022: 76; 2023: 58; 2024: 22; 2025: 11; 2026: 6.

205 papers.

Sort:

By area:

  1. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 397
  2. 2023A Survey on Self-supervised Learning: Algorithms, Applications, and Future Trendscited by 205
  3. 2022Multi-Granularity Cross-modal Alignment for Generalized Medical Visual Representation Learningcited by 159
  4. 2021Efficient Self-supervised Vision Transformers for Representation Learningcited by 149
  5. 2022Federated Learning from Pre-Trained Models: A Contrastive Learning Approachcited by 141
  6. 2022Revealing the Dark Secrets of Masked Image Modelingcited by 117
  7. 2022FreeSOLO: Learning to Segment Objects without Annotationscited by 102
  8. 2021Region Similarity Representation Learningcited by 96
  9. 2021Class-agnostic Object Detection with Multi-modal Transformercited by 85
  10. 2021Joint Learning of Localized Representations from Medical Images and Reportscited by 81
  11. 2021DETReg: Unsupervised Pretraining with Region Priors for Object Detectioncited by 76
  12. 2021Unsupervised Object-Level Representation Learning from Scene Imagescited by 72
38 more
  1. 2021Open Vocabulary Object Detection with Pseudo Bounding-Box Labelscited by 71
  2. 2022Self-Supervised Visual Representation Learning with Semantic Groupingcited by 70
  3. 2022Crafting Better Contrastive Views for Siamese Representation Learningcited by 67
  4. 2022CoCoNet: Coupled Contrastive Learning Network with Multi-level Feature Ensemble for Multi-modality Image Fusioncited by 63
  5. 2021Do Different Tracking Tasks Require Different Appearance Models?cited by 62
  6. 2022Patch-level Representation Learning for Self-supervised Vision Transformerscited by 60
  7. 2022Dual Temperature Helps Contrastive Learning Without Many Negative Samples: Towards Understanding and Simplifying MoCocited by 52
  8. 2022Understanding Masked Image Modeling via Learning Occlusion Invariant Featurecited by 50
  9. 2021Contrastive Learning for Neural Topic Modelcited by 48
  10. 2022DiRA: Discriminative, Restorative, and Adversarial Learning for Self-supervised Medical Image Analysiscited by 48
  11. 2023Towards In-context Scene Understandingcited by 45
  12. 2021Improving Contrastive Learning by Visualizing Feature Transformationcited by 44
  13. 2023HiCLIP: Contrastive Language-Image Pretraining with Hierarchy-aware Attentioncited by 44
  14. 2021SODA10M: A Large-Scale 2D Self/Semi-Supervised Object Detection Dataset for Autonomous Drivingcited by 43
  15. 2022HCSC: Hierarchical Contrastive Selective Codingcited by 43
  16. 2022ProposalCLIP: Unsupervised Open-Category Object Proposal Generation via Exploiting CLIP Cuescited by 42
  17. 2021MultiSiam: Self-supervised Multi-instance Siamese Representation Learning for Autonomous Drivingcited by 41
  18. 2022Point-Level Region Contrast for Object Detection Pre-Trainingcited by 41
  19. 2023Continual Vision-Language Representation Learning with Off-Diagonal Informationcited by 40
  20. 2022Mitigating Data Sparsity for Short Text Topic Modeling by Topic-Semantic Contrastive Learningcited by 38
  21. 2023DreamTeacher: Pretraining Image Backbones with Deep Generative Modelscited by 37
  22. 2021Revisiting the Transferability of Supervised Pretraining: an MLP Perspectivecited by 36
  23. 2021Rethinking Supervised Pre-training for Better Downstream Transferringcited by 35
  24. 2021INTERN: A New Learning Paradigm Towards General Visioncited by 33
  25. 2022Towards Discriminative Representation: Multi-view Trajectory Contrastive Learning for Online Multi-object Trackingcited by 33
  26. 2022Self-Supervised Learning via Maximum Entropy Codingcited by 33
  27. 2021Consistency-based Active Learning for Object Detectioncited by 32
  28. 2022Unsupervised Pre-training for Temporal Action Localization Taskscited by 31
  29. 2022Architecture-Agnostic Masked Image Modeling -- From ViT back to CNNcited by 31
  30. 2022One-Shot Transfer of Affordance Regions? AffCorrs!cited by 31
  31. 2022Towards All-in-one Pre-training via Maximizing Multi-modal Mutual Informationcited by 31
  32. 2022UniVIP: A Unified Framework for Self-Supervised Visual Pre-trainingcited by 30
  33. 2022Effective Adaptation in Multi-Task Co-Training for Unified Autonomous Drivingcited by 30
  34. 2023Contrastive Conditional Latent Diffusion for Audio-visual Segmentationcited by 30
  35. 2023Spherical Space Feature Decomposition for Guided Depth Map Super-Resolutioncited by 28
  36. 2023Small Object Detection via Coarse-to-fine Proposal Generation and Imitation Learningcited by 27
  37. 2021Semantics-Guided Contrastive Network for Zero-Shot Object detectioncited by 26
  38. 2024ConGeo: Robust Cross-view Geo-localization across Ground View Variationscited by 25

50 most cited of 205 All 205 (JSON)