Research map / Detection and segmentation / Weakly supervised detection and segmentation
Attention-Based Dropout Layer for Weakly Supervised Object Localization
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 |
|---|---|---|---|
| Add ADL | VGG-GAP · CUB-200-2011 | Top-1 Loc: 34.41 -> 52.36 | Helped |
| Add ADL | ResNet50-SE · CUB-200-2011 | Top-1 Loc: 42.72 -> 62.29 | Helped |
| Add ADL | InceptionV3 · ImageNet-1k | Top-1 Loc: 46.29* -> 48.71 | Helped |
| Add ADL | VGG-GAP · ImageNet-1k | Top-1 Loc: 45.83* -> 44.92 | Hurt |
| Use a 75% drop rate rather than applying the drop mask every iteration (ablation) | ADL on VGG-GAP · CUB-200-2011 | Top-1 Loc: 44.11 -> 49.69 | Helped |
| Set drop rate to 0% rather than 25% (ablation) | ADL on VGG-GAP · CUB-200-2011 | Top-1 Loc: 47.98 -> 32.24 | Hurt |
| Add an importance map alongside the drop mask (ablation) | ADL on VGG-GAP · CUB-200-2011 | Top-1 Loc: 47.67 -> 49.69 | Helped |
| Add a drop mask alongside the importance map (ablation) | ADL on VGG-GAP · CUB-200-2011 | Top-1 Loc: 33.91 -> 49.69 | Helped |
| Add ADL to pool4 and pool3 after conv5-3 (ablation) | ADL on VGG-GAP · CUB-200-2011 | Top-1 Loc: 41.73 -> 48.02; 48.02 -> 52.36 | Helped |
| Add ADL to lower-level pool2 after pool3 (ablation) | ADL on VGG-GAP · CUB-200-2011 | Top-1 Loc: 52.36 -> 48.46 | Hurt |
| Add ADL to pool1 after pool2 (ablation) | ADL on VGG-GAP · CUB-200-2011 | Top-1 Loc: 48.46 -> 49.69 | Helped |
Closest papers by meaning
- Evaluating Weakly Supervised Object Localization Methods Right (CVPR 2020 · similarity 0.91)
- Shallow Feature Matters for Weakly Supervised Object Localization (CVPR 2021 · similarity 0.90)
- Rethinking the Route Towards Weakly Supervised Object Localization (CVPR 2020 · similarity 0.90)
- Weakly Supervised Object Localization As Domain Adaption (CVPR 2022 · similarity 0.90)
- Learning to Localize Objects with Noisy Labeled Instances (AAAI 2019 · similarity 0.88)
- CAM Back Again: Large Kernel CNNs from a Weakly Supervised Object Localization Perspective (CVPR 2024 · similarity 0.84)
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.