Research map / Detection and segmentation / Weakly supervised detection and segmentation
Diffusion-Guided Weakly Supervised Semantic Segmentation
Sung-Hoon Yoon, Hoyong Kwon, Jaeseok Jeong, Daehee Park, Kuk-Jin Yoon
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use Min-Max K pooling instead of GAP for class prediction in ViT-CAM (ablation) | MCTformer-style ViT CAM · PASCAL VOC 2012 train (CAM seed) | mIoU: 63.3 -> 65.7 | Improved |
| add LFCA Diffusion-CAM supervision (Llfca) to ViT-CAMs (ablation) | DiG baseline · PASCAL VOC 2012 train (CAM seed) | mIoU: 65.7 -> 67.4 | Improved |
| add Patch Affinity Consistency (PAC) with denoised image (ablation) | DiG baseline · PASCAL VOC 2012 train (CAM seed) | mIoU: 65.7 -> 67.3 | Improved |
| combine LFCA and PAC (ablation) | DiG baseline · PASCAL VOC 2012 train (CAM seed) | mIoU: 65.7 -> 69.3 | Improved |
| add KL divergence distillation from ViT to LFCA (ablation) | DiG (LFCA + PAC) · PASCAL VOC 2012 train (CAM seed) | mIoU: 68.2 -> 69.3 | Improved |
| apply LFCA and PAC on top of GAP pooling (ablation) | MCTformer-style baseline with GAP · PASCAL VOC 2012 train (CAM seed) | mIoU: 63.3 -> 67.9 | Improved |
| use denoised image at t < 150 vs larger steps for PAC (ablation) | PAC · PASCAL VOC 2012 train (CAM seed) | mIoU: 67.3 ± 0.3% | Mixed |
| train segmentation model on DiG pseudo labels | DeepLab V1 WRN38 · PASCAL VOC 2012 val | mIoU: 73.3 -> 73.9 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- USAGE — USAGE: A Unified Seed Area Generation Paradigm for Weakly Supervised Semantic Segmentation (ICCV 2023) · better on 4 of 4 shared results
- MCTformer — MCTformer+: Multi-Class Token Transformer for Weakly Supervised Semantic Segmentation (TPAMI 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Phase Concentration and Shortcut Suppression for Weakly Supervised Semantic Segmentation (ECCV 2024 · similarity 0.91)
- Max Pooling with Vision Transformers Reconciles Class and Shape in Weakly Supervised Semantic Segmentation (ECCV 2022 · similarity 0.91)
- Weakly Supervised Semantic Segmentation via Progressive Confidence Region Expansion (CVPR 2025 · similarity 0.89)
- VidSeg: Training-free Video Semantic Segmentation based on Diffusion Models (CVPR 2025 · similarity 0.88)
- Progressive Feature Self-Reinforcement for Weakly Supervised Semantic Segmentation (AAAI 2024 · similarity 0.88)
- Weakly Supervised Semantic Segmentation for Driving Scenes (AAAI 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
