mlsanityData to 14 Oct 2026

Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentation

arXiv 1803.10464 ↗First posted 28 Mar 2018cs.CVcited by 321

Not on the map. Why?

Abstract (arXiv)

The deficiency of segmentation labels is one of the main obstacles to semantic segmentation in the wild. To alleviate this issue, we present a novel framework that generates segmentation labels of images given their image-level class labels. In this weakly supervised setting, trained models have been known to segment local discriminative parts rather than the entire object area. Our solution is to propagate such local responses to nearby areas which belong to the same semantic entity. To this end, we propose a Deep Neural Network (DNN) called AffinityNet that predicts semantic affinity between a pair of adjacent image coordinates. The semantic propagation is then realized by random walk with the affinities predicted by AffinityNet. More importantly, the supervision employed to train AffinityNet is given by the initial discriminative part segmentation, which is incomplete as a segmentation annotation but sufficient for learning semantic affinities within small image areas. Thus the entire framework relies only on image-level class labels and does not require any extra data or annotations. On the PASCAL VOC 2012 dataset, a DNN learned with segmentation labels generated by our method outperforms previous models trained with the same level of supervision, and is even as competitive as those relying on stronger supervision.

Key facts. Cited by 321 papers in the map (data through 14 Oct 2026). First posted on arXiv 28 Mar 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 71; 2022: 69; 2023: 75; 2024: 57; 2025: 27; 2026: 22.

321 papers.

Sort:

By area:

  1. 2022GroupViT: Semantic Segmentation Emerges from Text Supervisioncited by 531
  2. 2021Simultaneously Localize, Segment and Rank the Camouflaged Objectscited by 198
  3. 2023DiffuMask: Synthesizing Images with Pixel-level Annotations for Semantic Segmentation Using Diffusion Modelscited by 162
  4. 2022Multi-class Token Transformer for Weakly Supervised Semantic Segmentationcited by 138
  5. 2021Anti-Adversarially Manipulated Attributions for Weakly and Semi-Supervised Semantic Segmentationcited by 134
  6. 2021Railroad is not a Train: Saliency as Pseudo-pixel Supervision for Weakly Supervised Semantic Segmentationcited by 127
  7. 2021TS-CAM: Token Semantic Coupled Attention Map for Weakly Supervised Object Localizationcited by 123
  8. 2023DatasetDM: Synthesizing Data with Perception Annotations Using Diffusion Modelscited by 123
  9. 2021Interpretable Deep Learning: Interpretation, Interpretability, Trustworthiness, and Beyondcited by 118
  10. 2022Learning Affinity from Attention: End-to-End Weakly-Supervised Semantic Segmentation with Transformerscited by 118
  11. 2022CLIP is Also an Efficient Segmenter: A Text-Driven Approach for Weakly Supervised Semantic Segmentationcited by 118
  12. 2022ReCo: Retrieve and Co-segment for Zero-shot Transfercited by 117
38 more
  1. 2021Semi-supervised Semantic Segmentation with Directional Context-aware Consistencycited by 112
  2. 2021BBAM: Bounding Box Attribution Map for Weakly Supervised Semantic and Instance Segmentationcited by 105
  3. 2021Large-scale Unsupervised Semantic Segmentationcited by 101
  4. 2021Pixel Contrastive-Consistent Semi-Supervised Semantic Segmentationcited by 101
  5. 2023Learning Open-vocabulary Semantic Segmentation Models From Natural Language Supervisioncited by 99
  6. 2021Non-Salient Region Object Mining for Weakly Supervised Semantic Segmentationcited by 95
  7. 2021One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentationcited by 89
  8. 2022Self-supervised Image-specific Prototype Exploration for Weakly Supervised Semantic Segmentationcited by 87
  9. 2021Weakly Supervised Semantic Segmentation by Pixel-to-Prototype Contrastcited by 85
  10. 2021Leveraging Auxiliary Tasks with Affinity Learning for Weakly Supervised Semantic Segmentationcited by 77
  11. 2021Complementary Patch for Weakly Supervised Semantic Segmentationcited by 75
  12. 2022Class Re-Activation Maps for Weakly-Supervised Semantic Segmentationcited by 75
  13. 2022Cross Language Image Matching for Weakly Supervised Semantic Segmentationcited by 72
  14. 2021Context Decoupling Augmentation for Weakly Supervised Semantic Segmentationcited by 69
  15. 2022Scribble-Supervised LiDAR Semantic Segmentationcited by 69
  16. 2021Reducing Information Bottleneck for Weakly Supervised Semantic Segmentationcited by 65
  17. 2023Token Contrast for Weakly-Supervised Semantic Segmentationcited by 65
  18. 2022Regional Semantic Contrast and Aggregation for Weakly Supervised Semantic Segmentationcited by 64
  19. 2021Adaptive Early-Learning Correction for Segmentation from Noisy Annotationscited by 62
  20. 2021DiscoBox: Weakly Supervised Instance Segmentation and Semantic Correspondence from Box Supervisioncited by 59
  21. 2022Weakly Supervised Semantic Segmentation using Out-of-Distribution Datacited by 58
  22. 2022Semi-supervised Semantic Segmentation with Error Localization Networkcited by 57
  23. 2022L2G: A Simple Local-to-Global Knowledge Transfer Framework for Weakly Supervised Semantic Segmentationcited by 57
  24. 2024Diffusion Models and Representation Learning: A Surveycited by 54
  25. 2023Segment Anything Model (SAM) Enhanced Pseudo Labels for Weakly Supervised Semantic Segmentationcited by 53
  26. 2021Discriminative Region Suppression for Weakly-Supervised Semantic Segmentationcited by 50
  27. 2021Pseudo-mask Matters in Weakly-supervised Semantic Segmentationcited by 48
  28. 2021ZeroWaste Dataset: Towards Deformable Object Segmentation in Cluttered Scenescited by 47
  29. 2021Affinity Attention Graph Neural Network for Weakly Supervised Semantic Segmentationcited by 47
  30. 2021Unveiling the Potential of Structure Preserving for Weakly Supervised Object Localizationcited by 46
  31. 2022Fully Convolutional Change Detection Framework with Generative Adversarial Network for Unsupervised, Weakly Supervised and Regional Supervised Change Detectioncited by 46
  32. 2022A Survey on Label-efficient Deep Image Segmentation: Bridging the Gap between Weak Supervision and Dense Predictioncited by 46
  33. 2021Universal Weakly Supervised Segmentation by Pixel-to-Segment Contrastive Learningcited by 43
  34. 2021Multi-Layer Pseudo-Supervision for Histopathology Tissue Semantic Segmentation using Patch-level Classification Labelscited by 43
  35. 2022ACSeg: Adaptive Conceptualization for Unsupervised Semantic Segmentationcited by 42
  36. 2021Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentationcited by 41
  37. 2022Max Pooling with Vision Transformers reconciles class and shape in weakly supervised semantic segmentationcited by 41
  38. 2023CLIP in Medical Imaging: A Surveycited by 40

50 most cited of 321 All 321 (JSON)