mlsanityData to 7 Oct 2026

Importance Sampling CAMs for Weakly-Supervised Segmentation

arXiv 2203.12459 ↗First posted 23 Mar 2022cs.CV cs.LG eess.IVcited by 3

Not on the map. Why?

Abstract (arXiv)

Classification networks can be used to localize and segment objects in images by means of class activation maps (CAMs). However, without pixel-level annotations, classification networks are known to (1) mainly focus on discriminative regions, and (2) to produce diffuse CAMs without well-defined prediction contours. In this work, we approach both problems with two contributions for improving CAM learning. First, we incorporate importance sampling based on the class-wise probability mass function induced by the CAMs to produce stochastic image-level class predictions. This results in CAMs which activate over a larger extent of objects. Second, we formulate a feature similarity loss term which aims to match the prediction contours with edges in the image. As a third contribution, we conduct experiments on the PASCAL VOC 2012 benchmark dataset to demonstrate that these modifications significantly increase the performance in terms of contour accuracy, while being comparable to current state-of-the-art methods in terms of region similarity.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 40 of its 63 references that are in the map. First posted on arXiv 23 Mar 2022.

Builds on

40 of 63 references.

References · 40

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  4. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  5. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  6. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  7. 2020Image Segmentation Using Deep Learning: A Surveycited by 659
  8. 2014Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFscited by 552
  9. 2019Res2Net: A New Multi-scale Backbone Architecturecited by 468
  10. 2016ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentationcited by 330
  11. 2018Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentationcited by 321
  12. 2020Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentationcited by 299
28 more
  1. 2015What's the Point: Semantic Segmentation with Point Supervisioncited by 297
  2. 2020Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learningcited by 292
  3. 2019Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relationscited by 245
  4. 2015BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentationcited by 238
  5. 2020Causal Intervention for Weakly-Supervised Semantic Segmentationcited by 229
  6. 2016Simple Does It: Weakly Supervised Instance and Semantic Segmentationcited by 226
  7. 2016Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentationcited by 198
  8. 2015Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentationcited by 189
  9. 2016Wider or Deeper: Revisiting the ResNet Model for Visual Recognitioncited by 184
  10. 2019FickleNet: Weakly and Semi-supervised Semantic Image Segmentation using Stochastic Inferencecited by 162
  11. 2018Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentationcited by 145
  12. 2020Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentationcited by 144
  13. 2022Multi-class Token Transformer for Weakly Supervised Semantic Segmentationcited by 138
  14. 2022Learning Affinity from Attention: End-to-End Weakly-Supervised Semantic Segmentation with Transformerscited by 118
  15. 2018Learning random-walk label propagation for weakly-supervised semantic segmentationcited by 101
  16. 2015Constrained Convolutional Neural Networks for Weakly Supervised Segmentationcited by 100
  17. 2018Self-Erasing Network for Integral Object Attentioncited by 95
  18. 2022Self-supervised Image-specific Prototype Exploration for Weakly Supervised Semantic Segmentationcited by 87
  19. 2021Weakly Supervised Semantic Segmentation by Pixel-to-Prototype Contrastcited by 85
  20. 2018Weakly-Supervised Semantic Segmentation by Iteratively Mining Common Object Featurescited by 81
  21. 2018CIAN: Cross-Image Affinity Net for Weakly Supervised Semantic Segmentationcited by 80
  22. 2015STC: A Simple to Complex Framework for Weakly-supervised Semantic Segmentationcited by 78
  23. 2021Complementary Patch for Weakly Supervised Semantic Segmentationcited by 75
  24. 2022Class Re-Activation Maps for Weakly-Supervised Semantic Segmentationcited by 75
  25. 2021Context Decoupling Augmentation for Weakly Supervised Semantic Segmentationcited by 69
  26. 2021Reducing Information Bottleneck for Weakly Supervised Semantic Segmentationcited by 65
  27. 2019Self-Supervised Difference Detection for Weakly-Supervised Semantic Segmentationcited by 58
  28. 2021Pseudo-mask Matters in Weakly-supervised Semantic Segmentationcited by 48

Built on by

Cited by year

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

3 papers.

  1. 2023Masked Collaborative Contrast for Weakly Supervised Semantic Segmentationcited by 2
  2. 2025From Missing Pieces to Masterpieces: Image Completion with Context-Adaptive Diffusioncited by 2
  3. 2023High-fidelity Pseudo-labels for Boosting Weakly-Supervised Segmentationcited by 1