mlsanityData to 7 Oct 2026

Instance Localization for Self-supervised Detection Pretraining

arXiv 2102.08318 ↗First posted 16 Feb 2021cs.CVcited by 99

Not on the map. Why?

Abstract (arXiv)

Prior research on self-supervised learning has led to considerable progress on image classification, but often with degraded transfer performance on object detection. The objective of this paper is to advance self-supervised pretrained models specifically for object detection. Based on the inherent difference between classification and detection, we propose a new self-supervised pretext task, called instance localization. Image instances are pasted at various locations and scales onto background images. The pretext task is to predict the instance category given the composited images as well as the foreground bounding boxes. We show that integration of bounding boxes into pretraining promotes better task alignment and architecture alignment for transfer learning. In addition, we propose an augmentation method on the bounding boxes to further enhance the feature alignment. As a result, our model becomes weaker at Imagenet semantic classification but stronger at image patch localization, with an overall stronger pretrained model for object detection. Experimental results demonstrate that our approach yields state-of-the-art transfer learning results for object detection on PASCAL VOC and MSCOCO.

Key facts. Cited by 99 papers in the map (data through 7 Oct 2026). Builds on 39 of its 47 references that are in the map. First posted on arXiv 16 Feb 2021.

Builds on

39 of 47 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. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  10. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  11. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  12. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,239
27 more
  1. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,845
  2. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,282
  3. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
  4. 2017Rethinking Atrous Convolution for Semantic Image Segmentationcited by 2,154
  5. 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,838
  6. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,586
  7. 2016Context Encoders: Feature Learning by Inpaintingcited by 1,424
  8. 2016Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzlescited by 1,379
  9. 2020Big Self-Supervised Models are Strong Semi-Supervised Learnerscited by 1,375
  10. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  11. 2016Colorful Image Colorizationcited by 1,113
  12. 2015Unsupervised Visual Representation Learning by Context Predictioncited by 1,062
  13. 2020What Makes for Good Views for Contrastive Learning?cited by 843
  14. 2018Rethinking ImageNet Pre-trainingcited by 376
  15. 2020Rethinking Pre-training and Self-trainingcited by 352
  16. 2016Generative Visual Manipulation on the Natural Image Manifoldcited by 348
  17. 2017Objects that Soundcited by 254
  18. 2019Large Scale Adversarial Representation Learningcited by 238
  19. 2017Cut, Paste and Learn: Surprisingly Easy Synthesis for Instance Detectioncited by 209
  20. 2017Multi-task Self-Supervised Visual Learningcited by 200
  21. 2020SpeedNet: Learning the Speediness in Videoscited by 171
  22. 2016Learning Features by Watching Objects Movecited by 138
  23. 2020What makes instance discrimination good for transfer learning?cited by 111
  24. 2019InstaBoost: Boosting Instance Segmentation via Probability Map Guided Copy-Pastingcited by 95
  25. 2020Video Representation Learning with Visual Tempo Consistencycited by 57
  26. 2020Distilling Localization for Self-Supervised Representation Learningcited by 34
  27. 2019Revisiting Feature Alignment for One-stage Object Detectioncited by 14

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 21; 2022: 29; 2023: 24; 2024: 13; 2025: 7; 2026: 5.

99 papers.

Sort:

By area:

  1. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 397
  2. 2023A Cookbook of Self-Supervised Learningcited by 248
  3. 2023A Survey on Self-supervised Learning: Algorithms, Applications, and Future Trendscited by 205
  4. 2021SPViT: Enabling Faster Vision Transformers via Soft Token Pruningcited by 165
  5. 2022CroCo: Self-Supervised Pre-training for 3D Vision Tasks by Cross-View Completioncited by 158
  6. 2021Efficient Self-supervised Vision Transformers for Representation Learningcited by 149
  7. 2022VICRegL: Self-Supervised Learning of Local Visual Featurescited by 118
  8. 2021MST: Masked Self-Supervised Transformer for Visual Representationcited by 114
  9. 2021Aligning Pretraining for Detection via Object-Level Contrastive Learningcited by 110
  10. 2021Region Similarity Representation Learningcited by 96
  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. 2022Self-Supervised Visual Representation Learning with Semantic Groupingcited by 70
  2. 2021Data-Efficient Instance Generation from Instance Discriminationcited by 58
  3. 2023MC-JEPA: A Joint-Embedding Predictive Architecture for Self-Supervised Learning of Motion and Content Featurescited by 53
  4. 2021Revisiting Contrastive Methods for Unsupervised Learning of Visual Representationscited by 48
  5. 2022HeatViT: Hardware-Efficient Adaptive Token Pruning for Vision Transformerscited by 46
  6. 2021Motion-aware Contrastive Video Representation Learning via Foreground-background Mergingcited by 45
  7. 2021Improving Contrastive Learning by Visualizing Feature Transformationcited by 44
  8. 2022Point-Level Region Contrast for Object Detection Pre-Trainingcited by 41
  9. 2022Learning to Drive by Watching YouTube Videos: Action-Conditioned Contrastive Policy Pretrainingcited by 37
  10. 2021A Survey of Self-Supervised and Few-Shot Object Detectioncited by 35
  11. 2022Towards Discriminative Representation: Multi-view Trajectory Contrastive Learning for Online Multi-object Trackingcited by 33
  12. 2022Unsupervised Pre-training for Temporal Action Localization Taskscited by 31
  13. 2022UniVIP: A Unified Framework for Self-Supervised Visual Pre-trainingcited by 30
  14. 2023PMatch: Paired Masked Image Modeling for Dense Geometric Matchingcited by 29
  15. 2023L-DAWA: Layer-wise Divergence Aware Weight Aggregation in Federated Self-Supervised Visual Representation Learningcited by 29
  16. 2022Your Contrastive Learning Is Secretly Doing Stochastic Neighbor Embeddingcited by 28
  17. 2022CP2: Copy-Paste Contrastive Pretraining for Semantic Segmentationcited by 26
  18. 2021Self-supervised Contrastive Learning of Multi-view Facial Expressionscited by 23
  19. 2021Object-Aware Cropping for Self-Supervised Learningcited by 22
  20. 2021Contextualized Spatio-Temporal Contrastive Learning with Self-Supervisioncited by 15
  21. 2023GridCLIP: One-Stage Object Detection by Grid-Level CLIP Representation Learningcited by 15
  22. 2023AlignDet: Aligning Pre-training and Fine-tuning in Object Detectioncited by 14
  23. 2022Spatial-then-Temporal Self-Supervised Learning for Video Correspondencecited by 13
  24. 2023R-MAE: Regions Meet Masked Autoencoderscited by 13
  25. 2024Representing Part-Whole Hierarchies in Foundation Models by Learning Localizability, Composability, and Decomposability from Anatomy via Self-Supervisioncited by 12
  26. 2026INSID3: Training-Free In-Context Segmentation with DINOv3cited by 12
  27. 2021RegionCL: Can Simple Region Swapping Contribute to Contrastive Learning?cited by 11
  28. 2022ConCL: Concept Contrastive Learning for Dense Prediction Pre-training in Pathology Imagescited by 11
  29. 2023Correlational Image Modeling for Self-Supervised Visual Pre-Trainingcited by 11
  30. 2021Deeply Unsupervised Patch Re-Identification for Pre-training Object Detectorscited by 9
  31. 2021Exploring Localization for Self-supervised Fine-grained Contrastive Learningcited by 9
  32. 2023Self-Supervised Representation Learning from Temporal Ordering of Automated Driving Sequencescited by 9
  33. 2022Dense Siamese Network for Dense Unsupervised Learningcited by 8
  34. 2022Unseen Object Instance Segmentation with Fully Test-time RGB-D Embeddings Adaptationcited by 8
  35. 2022Self-supervised Learning with Local Contrastive Loss for Detection and Semantic Segmentationcited by 8
  36. 2022Global-Local Self-Distillation for Visual Representation Learningcited by 8
  37. 2025SANSA: Unleashing the Hidden Semantics in SAM2 for Few-Shot Segmentationcited by 8
  38. 2023Understanding Self-Supervised Features for Learning Unsupervised Instance Segmentationcited by 7

50 most cited of 99 All 99 (JSON)