Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly Supervised Semantic Segmentation via Adversarial Learning of Classifier and Reconstructor
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| adversarial training of classifier and reconstructor instead of frozen pre-trained reconstructor (ablation) | CAM classifier with reconstructor · PASCAL VOC 2012 train | mIoU (%): +2.9 (52.9 to 55.8) | Helped |
| adding Stochastic Remnant Feeding (SRF) to the reconstructor (ablation) | adversarial and pre-trained reconstructor settings · PASCAL VOC 2012 train | mIoU (%): +4.5 (55.8 to 60.3) adversarial; +1.7 (52.9 to 54.6) pre-trained | Helped |
| adding the reconstruction-based target and non-target losses to the classifier loss (ablation) | CAM classifier · PASCAL VOC 2012 train | mIoU (%): +11.9 (48.4 to 60.3) over classification-only baseline | Helped |
| removing non-target loss term from classifier objective (ablation) | CAM classifier · PASCAL VOC 2012 train | mIoU (%): -6.9 (60.3 to 53.4) | Hurt |
| removing target loss term from classifier objective (ablation) | CAM classifier · PASCAL VOC 2012 train | mIoU (%): -8.8 (60.3 to 51.5) | Hurt |
| adversarial classifier-reconstructor framework (ACR) vs AE-based AEFT | WSSS pipeline with WRN38 · PASCAL VOC 2012 val | mIoU (%): +1.0 (70.9 to 71.9) | Helped |
| ACR with ViT-based CAM refinement | ACR pipeline · PASCAL VOC 2012 val | mIoU (%): +0.5 (71.9 to 72.4) | Helped |
Closest papers by meaning
- Combining Bottom-Up, Top-Down, and Smoothness Cues for Weakly Supervised Image Segmentation (CVPR 2017 · similarity 0.89)
- Repurposing GANs for One-Shot Semantic Part Segmentation (CVPR 2021 · similarity 0.88)
- Progressive Feature Self-Reinforcement for Weakly Supervised Semantic Segmentation (AAAI 2024 · similarity 0.88)
- Class Tokens Infusion for Weakly Supervised Semantic Segmentation (CVPR 2024 · similarity 0.88)
- Separate and Conquer: Decoupling Co-occurrence via Decomposition and Representation for Weakly Supervised Semantic Segmentation (CVPR 2024 · similarity 0.87)
- Weakly Supervised Semantic Segmentation via Progressive Confidence Region Expansion (CVPR 2025 · similarity 0.87)
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.