mlsanityData to 7 Oct 2026

Weak-shot Semantic Segmentation by Transferring Semantic Affinity and Boundary

arXiv 2110.01519 ↗First posted 4 Oct 2021cs.CVcited by 7

Not on the map. Why?

Abstract (arXiv)

Weakly-supervised semantic segmentation (WSSS) with image-level labels has been widely studied to relieve the annotation burden of the traditional segmentation task. In this paper, we show that existing fully-annotated base categories can help segment objects of novel categories with only image-level labels, even if base categories and novel categories have no overlap. We refer to this task as weak-shot semantic segmentation, which could also be treated as WSSS with auxiliary fully-annotated categories. Recent advanced WSSS methods usually obtain class activation maps (CAMs) and refine them by affinity propagation. Based on the observation that semantic affinity and boundary are class-agnostic, we propose a method under the WSSS framework to transfer semantic affinity and boundary from base to novel categories. As a result, we find that pixel-level annotation of base categories can facilitate affinity learning and propagation, leading to higher-quality CAMs of novel categories. Extensive experiments on PASCAL VOC 2012 dataset prove that our method significantly outperforms WSSS baselines on novel categories.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 46 of its 63 references that are in the map. First posted on arXiv 4 Oct 2021.

Builds on

46 of 63 references.

References · 46

  1. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  2. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,239
  3. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  4. 2015Multi-Scale Context Aggregation by Dilated Convolutionscited by 1,128
  5. 2014Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFscited by 552
  6. 2019PANet: Few-Shot Image Semantic Segmentation with Prototype Alignmentcited by 449
  7. 2020Semi-Supervised Semantic Segmentation with Cross-Consistency Trainingcited by 375
  8. 2015Learning Deconvolution Network for Semantic Segmentationcited by 365
  9. 2017One-Shot Learning for Semantic Segmentationcited by 337
  10. 2018Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentationcited by 321
  11. 2020Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentationcited by 299
  12. 2019Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relationscited by 245
34 more
  1. 2019Zero-Shot Semantic Segmentationcited by 245
  2. 2020Causal Intervention for Weakly-Supervised Semantic Segmentationcited by 229
  3. 2017Hide-and-Seek: Forcing a Network to be Meticulous for Weakly-supervised Object and Action Localizationcited by 207
  4. 2017Object Region Mining with Adversarial Erasing: A Simple Classification to Semantic Segmentation Approachcited by 201
  5. 2016Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentationcited by 198
  6. 2015Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentationcited by 189
  7. 2016Wider or Deeper: Revisiting the ResNet Model for Visual Recognitioncited by 184
  8. 2018Adversarial Learning for Semi-Supervised Semantic Segmentationcited by 173
  9. 2018SG-One: Similarity Guidance Network for One-Shot Semantic Segmentationcited by 169
  10. 2020Single-Stage Semantic Segmentation from Image Labelscited by 167
  11. 2019FickleNet: Weakly and Semi-supervised Semantic Image Segmentation using Stochastic Inferencecited by 162
  12. 2018Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentationcited by 145
  13. 2020Weakly-Supervised Semantic Segmentation via Sub-category Explorationcited by 145
  14. 2020Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentationcited by 144
  15. 2021Anti-Adversarially Manipulated Attributions for Weakly and Semi-Supervised Semantic Segmentationcited by 134
  16. 2021Deep Occlusion-Aware Instance Segmentation with Overlapping BiLayerscited by 104
  17. 2020Context-aware Feature Generation for Zero-shot Semantic Segmentationcited by 94
  18. 2019Reliability Does Matter: An End-to-End Weakly Supervised Semantic Segmentation Approachcited by 88
  19. 2018Weakly-Supervised Semantic Segmentation by Iteratively Mining Common Object Featurescited by 81
  20. 2021Complementary Patch for Weakly Supervised Semantic Segmentationcited by 75
  21. 2015Efficient piecewise training of deep structured models for semantic segmentationcited by 71
  22. 2017Learning to Segment Every Thingcited by 64
  23. 2019Self-Supervised Difference Detection for Weakly-Supervised Semantic Segmentationcited by 58
  24. 2019ShapeMask: Learning to Segment Novel Objects by Refining Shape Priorscited by 42
  25. 2018Semi-Supervised Semantic Image Segmentation with Self-correcting Networkscited by 36
  26. 2018Few-Shot Segmentation Propagation with Guided Networkscited by 29
  27. 2020Boosting Weakly Supervised Object Detection with Progressive Knowledge Transfercited by 28
  28. 2015Decoupled Deep Neural Network for Semi-supervised Semantic Segmentationcited by 23
  29. 2016Convolutional Random Walk Networks for Semantic Image Segmentationcited by 20
  30. 2020Weak-shot Fine-grained Classification via Similarity Transfercited by 18
  31. 2018Mixed Supervised Object Detection with Robust Objectness Transfercited by 14
  32. 2020Commonality-Parsing Network across Shape and Appearance for Partially Supervised Instance Segmentationcited by 14
  33. 2017Semi and Weakly Supervised Semantic Segmentation Using Generative Adversarial Networkcited by 11
  34. 2020Cross-Supervised Object Detectioncited by 3

Built on by

Cited by year

Citing papers per year: 2021: 3; 2022: 2; 2023: 1; 2024: 0; 2025: 1.

7 papers.

  1. 2021Novel Class Discovery in Semantic Segmentationcited by 31
  2. 2021Mixed Supervised Object Detection by Transferring Mask Prior and Semantic Similaritycited by 11
  3. 2022Weak-shot Semantic Segmentation via Dual Similarity Transfercited by 10
  4. 2022Lesion Guided Explainable Few Weak-shot Medical Report Generationcited by 6
  5. 2023Learning Cross-Modal Affinity for Referring Video Object Segmentation Targeting Limited Samplescited by 6
  6. 2025Few-Shot Referring Video Single- and Multi-Object Segmentation via Cross-Modal Affinity with Instance Sequence Matchingcited by 5
  7. 2021Weak Novel Categories without Tears: A Survey on Weak-Shot Learningcited by 0