Research map / Detection and segmentation / Weakly supervised detection and segmentation
Pro2SAM: Mask Prompt to SAM with Grid Points for Weakly Supervised Object Localization
Xi Yang, Songsong Duan, Nannan Wang, Xinbo Gao
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 |
|---|---|---|---|
| replace self-attention map with global tokens in GTFormer (ablation) | GTFormer · ILSVRC | Top-1 Loc: 43.0 -> 60.2 | Improved |
| add global-token attention (GTA) (ablation) | GTFormer · ILSVRC | Top-1 Loc: 51.1 -> 60.2 | Improved |
| use 4 global tokens instead of 1 (ablation) | GTFormer · ILSVRC | Top-1 Loc: 59.4 -> 60.2 | Improved |
| use 16 global tokens instead of 4 (ablation) | GTFormer · ILSVRC | Top-1 Loc: 60.2 -> 41.3 | Hurt |
| set loss weights mu:lambda to 1:0.5 instead of 1:1 (ablation) | GTFormer · ILSVRC | Top-1 Loc: 43.6 -> 60.2 | Improved |
| mask matching instead of bounding box matching of SAM masks (ablation) | Pro2SAM · ILSVRC | IOU50: 53.3 -> 77.8 | Improved |
| mask matching instead of combined mask and bbox matching (ablation) | Pro2SAM · ILSVRC | IOU90: 40.3 -> 45.9 | Improved |
| add SAM grid-point mask matching on top of GTFormer | GTFormer · ILSVRC | Top-1 Loc: 60.2 -> 66.85 | Improved |
| add SAM on top of GTFormer | GTFormer · CUB-200-2011 | GT-Known: 98.48 -> 95.67 | Hurt |
| add SAM on top of GTFormer, fine-grained localization | GTFormer · CUB-200-2011 | MaxBoxAccV2 IOU90: 8.70 -> 48.43 | 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.
- C2AM — C2AM: Contrastive Learning of Class-Agnostic Activation Map for Weakly Supervised Object Localization and Semantic Segmentation (CVPR 2022) · better on 2 of 2 shared results
- TokenCut — TokenCut: Segmenting Objects in Images and Videos With Self-Supervised Transformer and Normalized Cut (TPAMI 2023) · better on 1 of 1 shared result
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)
- Evaluating Weakly Supervised Object Localization Methods Right (CVPR 2020 · similarity 0.91)
- Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets (TPAMI 2022 · similarity 0.91)
- Rethinking Class Activation Mapping for Weakly Supervised Object Localization (ECCV 2020 · similarity 0.90)
- Spatial-Aware Token for Weakly Supervised Object Localization (ICCV 2023 · similarity 0.90)
- CLIP-Driven Transformer for Weakly Supervised Object Localization (TPAMI 2025 · similarity 0.90)
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.
