Research map / Detection and segmentation / Weakly supervised detection and segmentation
Instance-Aware, Context-Focused, and Memory-Efficient Weakly Supervised Object Detection
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 MIST pseudo-box selection with spatial diversity and instance-associative constraints (ablation) | WSOD baseline · PASCAL VOC 2007 test | AP50: 42.5 -> 48.3 | Helped |
| Add ROI regression to MIST (ablation) | MIST WSOD · PASCAL VOC 2007 test | AP50: 48.3 -> 51.4 | Helped |
| Add learnable Concrete DropBlock (ablation) | MIST WSOD · PASCAL VOC 2007 test | AP50: 51.4 -> 54.9 | Helped |
| Use image-level spatial dropout instead of Concrete DropBlock (ablation) | MIST WSOD · PASCAL VOC 2007 test | AP50: 54.9 -> 51.1 | Hurt |
| Use independent spatial dropout on each ROI instead of Concrete DropBlock (ablation) | MIST WSOD · PASCAL VOC 2007 test | AP50: 54.9 -> 52.4 | Hurt |
| Use non-parametric DropBlock instead of Concrete DropBlock (ablation) | MIST WSOD · PASCAL VOC 2007 test | AP50: 54.9 -> 52.9 | Hurt |
| Add MIST (ablation) | WSOD baseline · PASCAL VOC 2007 | AR100: 32.5 -> 43.9 | Helped |
| Train with the larger VOC 2007+2012 dataset rather than VOC 2007 alone | Ours · PASCAL VOC 2007 test | AP50: 54.9 -> 58.1 | Helped |
| Train on VOC 2012 rather than VOC 2007 | Ours · PASCAL VOC 2007 test | AP50: 54.9 -> 56.3 | Helped |
| Add optical flow | Video WSOD · ImageNet VID | Det. (AP): 45.7 -> 46.9 | Helped |
Closest papers by meaning
- W2F: A Weakly-Supervised to Fully-Supervised Framework for Object Detection (CVPR 2018 · similarity 0.91)
- Multiple Instance Detection Network With Online Instance Classifier Refinement (CVPR 2017 · similarity 0.91)
- Object Instance Mining for Weakly Supervised Object Detection (AAAI 2020 · similarity 0.90)
- You Reap What You Sow: Using Videos to Generate High Precision Object Proposals for Weakly-Supervised Object Detection (CVPR 2019 · similarity 0.90)
- Instance Mining with Class Feature Banks for Weakly Supervised Object Detection (AAAI 2021 · similarity 0.89)
- Weak-Shot Object Detection Through Mutual Knowledge Transfer (CVPR 2023 · 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.