Research map / Detection and segmentation / Weakly supervised detection and segmentation
TS-CAM: Token Semantic Coupled Attention Map for Weakly Supervised Object Localization
Wei Gao, Fang Wan, Xingjia Pan, Zhiliang Peng, Qi Tian, Zhenjun Han, Bolei Zhou, Qixiang Ye
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 |
|---|---|---|---|
| add semantic-aware map to attention (couple them) (ablation) | TransAttention · CUB-200-2011 | Top-1 Loc Acc: 58.9 -> 71.3 | Improved |
| add attention map to semantic-aware map (couple them) (ablation) | TransCAM · CUB-200-2011 | Top-1 Loc Acc: 17.7 -> 71.3 | Improved |
| add semantic-aware map to attention (couple them) (ablation) | TransAttention · ILSVRC | Top-1 Loc Acc: 43.0 -> 53.4 | Improved |
| add attention to semantic-aware map (ablation) | TransCAM · ILSVRC | Top-1 Loc Acc: 34.9 -> 53.4 | Improved |
| use Conv2D instead of FC for semantic-aware map (ablation) | TS-CAM · ILSVRC | Top-1 Loc Acc: 31.5 -> 53.4 | Improved |
| use Conv2D instead of FC (ablation) | TS-CAM · CUB-200-2011 | Top-1 Loc Acc: 69.4 -> 71.3 | Improved |
| use Conv2D instead of Conv1D (ablation) | TS-CAM · ILSVRC | Top-1 Loc Acc: 49.8 -> 53.4 | Improved |
| sum attention maps from all 12 layers instead of 8 (ablation) | TS-CAM · CUB-200-2011 | Top-1 Loc Acc: 65.2 -> 71.3 | Improved |
| sum attention maps from all 12 layers instead of 8 (ablation) | TS-CAM · ILSVRC | Top-1 Loc Acc: 49.4 -> 53.4 | Improved |
| replace CNN CAM with TS-CAM on DeiT-S | CAM (VGG16) · CUB-200-2011 | Top-1 Loc Acc: 44.2 -> 71.3 | Improved |
| replace CAM InceptionV3 with TS-CAM | CAM (InceptionV3) · ILSVRC | Top-1 Loc Acc: 46.3 -> 53.4 | Improved |
| replace CAM InceptionV3 with TS-CAM | CAM (InceptionV3) · CUB-200-2011 | Part error: 23.09 -> 6.30 | Improved |
Later papers that compared with this method
9 later accepted papers report TS-CAM as a baseline in their results tables; 9 of them report a better number on most of the results they share with it.
- Cross-token Guidance Transformer for Weakly Supervised Object Localization (ECCV 2026 · better on 5 of 5 shared results)
- Coarse2Fine: Local Consistency Aware Re-prediction for Weakly Supervised Object Localization (AAAI 2023 · better on 3 of 3 shared results)
- Generative Prompt Model for Weakly Supervised Object Localization (ICCV 2023 · better on 2 of 2 shared results)
- Category-aware Allocation Transformer for Weakly Supervised Object Localization (ICCV 2023 · better on 2 of 2 shared results)
- Weakly Supervised Object Localization via Transformer with Implicit Spatial Calibration (ECCV 2022 · better on 10 of 12 shared results)
- LCTR: On Awakening the Local Continuity of Transformer for Weakly Supervised Object Localization (AAAI 2022 · better on 6 of 6 shared results)
- Weakly Supervised Object Localization through Inter-class Feature Similarity and Intra-Class Appearance Consistency (ECCV 2022 · better on 3 of 3 shared results)
- Multi-Class Token Transformer for Weakly Supervised Semantic Segmentation (CVPR 2022 · better on 2 of 2 shared results)
1 more not listed.
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
- Weakly Supervised Object Localization via Transformer with Implicit Spatial Calibration (ECCV 2022 · similarity 0.91)
- Spatial-Aware Token for Weakly Supervised Object Localization (ICCV 2023 · similarity 0.89)
- Attention Guided CAM: Visual Explanations of Vision Transformer Guided by Self-Attention (AAAI 2024 · similarity 0.89)
- Cross-token Guidance Transformer for Weakly Supervised Object Localization (ECCV 2026 · similarity 0.89)
- MoRe: Class Patch Attention Needs Regularization for Weakly Supervised Semantic Segmentation (AAAI 2025 · similarity 0.89)
- CAM Back Again: Large Kernel CNNs from a Weakly Supervised Object Localization Perspective (CVPR 2024 · similarity 0.89)
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.
