Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly Supervised Object Localization As Domain Adaption
Check 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 |
|---|---|---|---|
| apply DA-WSOL with MMD to CAM | CAM with ResNet50 · OpenImages | pIoU: 42.95 -> 49.68 | Helped |
| apply DA-WSOL with MMD to CAM | CAM with ResNet50 · OpenImages | PxAP: 58.19 -> 65.42 | Helped |
| add MMD alignment without Universum regularization or TSA (ablation) | CAM with ResNet50 · OpenImages | pIoU: 42.95 -> 42.88 | Hurt |
| add MMD alignment without Universum regularization or TSA (ablation) | CAM with ResNet50 · OpenImages | PxAP: 58.19 -> 57.39 | Hurt |
| add TSA to MMD alignment without Universum regularization (ablation) | CAM with ResNet50 · OpenImages | pIoU: 42.88 -> 43.24 | Helped |
| add Universum regularization and TSA without MMD (ablation) | CAM with ResNet50 · OpenImages | pIoU: 42.95 -> 45.17 | Helped |
| add Universum regularization and TSA without MMD (ablation) | CAM with ResNet50 · OpenImages | PxAP: 58.19 -> 61.50 | Helped |
| add the full DAL loss, including MMD, Universum regularization, and TSA (ablation) | CAM with ResNet50 · OpenImages | pIoU: 42.95 -> 49.68 | Helped |
| add the full DAL loss, including MMD, Universum regularization, and TSA (ablation) | CAM with ResNet50 · OpenImages | PxAP: 58.19 -> 65.42 | Helped |
| add a classification stage | DA-WSOL with ResNet50 · ImageNet | Top-1 Loc: 43.26 -> 55.84 | Helped |
| use DANN instead of MMD for the DA component | CAM with ResNet50 · OpenImages | pIoU: 42.95 -> 46.71 | Helped |
| use DA-WSOL with MMD | CAM, HAS, CutMix, and ADL · OpenImages | pIoU: 42.95 -> 49.68; 41.92 -> 49.25; 42.73 -> 49.32; 42.05 -> 43.40 | Helped |
Closest papers by meaning
- Evaluating Weakly Supervised Object Localization Methods Right (CVPR 2020 · similarity 0.94)
- Rethinking the Route Towards Weakly Supervised Object Localization (CVPR 2020 · similarity 0.92)
- Strengthen Learning Tolerance for Weakly Supervised Object Localization (CVPR 2021 · similarity 0.91)
- Shallow Feature Matters for Weakly Supervised Object Localization (CVPR 2021 · similarity 0.91)
- CREAM: Weakly Supervised Object Localization via Class RE-Activation Mapping (CVPR 2022 · similarity 0.90)
- CAM Back Again: Large Kernel CNNs from a Weakly Supervised Object Localization Perspective (CVPR 2024 · similarity 0.85)
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.