Localized Region Guidance for Class Activation Mapping in WSSS
Not on the map. Why?
Abstract (arXiv)
Weakly Supervised Semantic Segmentation (WSSS) addresses the challenge of training segmentation models using only image-level annotations. Existing WSSS methods struggle with precise object boundary localization and focus only on the most discriminative regions. To address these challenges, we propose IG-CAM (Instance-Guided Class Activation Mapping), a novel approach that leverages instance-level cues and influence functions to generate high-quality, boundary-aware localization maps. Our method introduces three key innovations: (1) Instance-Guided Refinement using object proposals to guide CAM generation, ensuring complete object coverage; (2) Influence Function Integration that captures the relationship between training samples and model predictions; and (3) Multi-Scale Boundary Enhancement with progressive refinement strategies. IG-CAM achieves state-of-the-art performance on PASCAL VOC 2012 with 82.3% mIoU before post-processing, improving to 86.6% after CRF refinement, significantly outperforming previous WSSS methods. Extensive ablation studies validate each component's contribution, establishing IG-CAM as a new benchmark for weakly supervised semantic segmentation.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 25 of its 43 references that are in the map. First posted on arXiv 15 Sep 2025.
Builds on
25 of 43 references.
References · 25
- 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
- 2015Learning Deep Features for Discriminative Localizationcited by 2,436
- 2017Understanding Black-box Predictions via Influence Functionscited by 1,741
- 2019Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networkscited by 582
- 2020Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentationcited by 300
- 2020Influence Functions in Deep Learning Are Fragilecited by 252
- 2019Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relationscited by 247
- 2020Causal Intervention for Weakly-Supervised Semantic Segmentationcited by 229
- 2019FickleNet: Weakly and Semi-supervised Semantic Image Segmentation using Stochastic Inferencecited by 162
- 2020Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNscited by 151
- 2021Anti-Adversarially Manipulated Attributions for Weakly and Semi-Supervised Semantic Segmentationcited by 134
- 2015Constrained Convolutional Neural Networks for Weakly Supervised Segmentationcited by 100
13 more
- 2019Reliability Does Matter: An End-to-End Weakly Supervised Semantic Segmentation Approachcited by 88
- 2022Self-supervised Image-specific Prototype Exploration for Weakly Supervised Semantic Segmentationcited by 87
- 2021Weakly Supervised Semantic Segmentation by Pixel-to-Prototype Contrastcited by 85
- 2018CIAN: Cross-Image Affinity Net for Weakly Supervised Semantic Segmentationcited by 80
- 2021Complementary Patch for Weakly Supervised Semantic Segmentationcited by 75
- 2022Class Re-Activation Maps for Weakly-Supervised Semantic Segmentationcited by 75
- 2018Interpreting Black Box Predictions using Fisher Kernelscited by 48
- 2021Universal Weakly Supervised Segmentation by Pixel-to-Segment Contrastive Learningcited by 43
- 2022Threshold Matters in WSSS: Manipulating the Activation for the Robust and Accurate Segmentation Model Against Thresholdscited by 37
- 2022Expansion and Shrinkage of Localization for Weakly-Supervised Semantic Segmentationcited by 37
- 2023Extracting Class Activation Maps from Non-Discriminative Features as wellcited by 30
- 2020Mixup-CAM: Weakly-supervised Semantic Segmentation via Uncertainty Regularizationcited by 24
- 2024SFC: Shared Feature Calibration in Weakly Supervised Semantic Segmentationcited by 19
Built on by
Citing papers per year: 2025: 1.
1 paper.