mlsanityData to 14 Oct 2026

Semantics-Consistent Feature Search for Self-Supervised Visual Representation Learning

arXiv 2212.06486 ↗First posted 13 Dec 2022cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

In contrastive self-supervised learning, the common way to learn discriminative representation is to pull different augmented "views" of the same image closer while pushing all other images further apart, which has been proven to be effective. However, it is unavoidable to construct undesirable views containing different semantic concepts during the augmentation procedure. It would damage the semantic consistency of representation to pull these augmentations closer in the feature space indiscriminately. In this study, we introduce feature-level augmentation and propose a novel semantics-consistent feature search (SCFS) method to mitigate this negative effect. The main idea of SCFS is to adaptively search semantics-consistent features to enhance the contrast between semantics-consistent regions in different augmentations. Thus, the trained model can learn to focus on meaningful object regions, improving the semantic representation ability. Extensive experiments conducted on different datasets and tasks demonstrate that SCFS effectively improves the performance of self-supervised learning and achieves state-of-the-art performance on different downstream tasks.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 42 of its 53 references that are in the map. First posted on arXiv 13 Dec 2022.

Builds on

42 of 53 references.

References · 42

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  4. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  5. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  6. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  7. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  8. 2017Mask R-CNNcited by 7,277
  9. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  10. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  11. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  12. 2020Exploring Simple Siamese Representation Learningcited by 2,866
30 more
  1. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
  2. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,375
  3. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,287
  4. 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,845
  5. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
  6. 2013Fine-Grained Visual Classification of Aircraftcited by 1,726
  7. 2021An Empirical Study of Training Self-Supervised Vision Transformerscited by 1,616
  8. 2020Big Self-Supervised Models are Strong Semi-Supervised Learnerscited by 1,380
  9. 2021VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learningcited by 1,057
  10. 2020Dense Contrastive Learning for Self-Supervised Visual Pre-Trainingcited by 512
  11. 2019S4L: Self-Supervised Semi-Supervised Learningcited by 335
  12. 2021With a Little Help from My Friends: Nearest-Neighbor Contrastive Learning of Visual Representationscited by 326
  13. 2020Propagate Yourself: Exploring Pixel-Level Consistency for Unsupervised Visual Representation Learningcited by 297
  14. 2021DetCo: Unsupervised Contrastive Learning for Object Detectioncited by 205
  15. 2021Decoupled Contrastive Learningcited by 143
  16. 2021Aligning Pretraining for Detection via Object-Level Contrastive Learningcited by 110
  17. 2021Contrastive Learning with Stronger Augmentationscited by 100
  18. 2021Region Similarity Representation Learningcited by 96
  19. 2020AdCo: Adversarial Contrast for Efficient Learning of Unsupervised Representations from Self-Trained Negative Adversariescited by 90
  20. 2021Unsupervised Object-Level Representation Learning from Scene Imagescited by 72
  21. 2022Crafting Better Contrastive Views for Siamese Representation Learningcited by 67
  22. 2020OBoW: Online Bag-of-Visual-Words Generation for Self-Supervised Learningcited by 65
  23. 2021Mean Shift for Self-Supervised Learningcited by 65
  24. 2021Spatially Consistent Representation Learningcited by 64
  25. 2022HCSC: Hierarchical Contrastive Selective Codingcited by 43
  26. 2020ISD: Self-Supervised Learning by Iterative Similarity Distillationcited by 33
  27. 2022UniVIP: A Unified Framework for Self-Supervised Visual Pre-trainingcited by 31
  28. 2021Exploring Set Similarity for Dense Self-supervised Representation Learningcited by 29
  29. 2021Dense Semantic Contrast for Self-Supervised Visual Representation Learningcited by 24
  30. 2021Momentum^2 Teacher: Momentum Teacher with Momentum Statistics for Self-Supervised Learningcited by 14

Built on by

Cited by year

Citing papers per year: 2023: 3; 2024: 4; 2025: 1.

8 papers.

  1. 2024LeOCLR: Leveraging Original Images for Contrastive Learning of Visual Representationscited by 3
  2. 2024Multi-Token Enhancing for Vision Representation Learningcited by 3
  3. 2023Semantic Positive Pairs for Enhancing Visual Representation Learning of Instance Discrimination Methodscited by 2
  4. 2023Enhancing Representations through Heterogeneous Self-Supervised Learningcited by 2
  5. 2023Semantic-Aware Autoregressive Image Modeling for Visual Representation Learningcited by 2
  6. 2024Learning Discriminative Spatio-temporal Representations for Semi-supervised Action Recognitioncited by 2
  7. 2024POA: Pre-training Once for Models of All Sizescited by 1
  8. 2025A Contrastive Learning Foundation Model Based on Perfectly Aligned Sample Pairs for Remote Sensing Imagescited by 1