Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly Supervised Object Localization With Progressive Domain Adaptation
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 MIL mining to class-specific proposals (ablation) | Progressive domain adaptation · PASCAL VOC 2007 trainval | CorLoc (average): 31.8 -> 41.2 | Helped |
| Add detection-network fine-tuning after MIL (ablation) | Progressive domain adaptation with AlexNet · PASCAL VOC 2007 trainval | CorLoc (average): 41.2 -> 49.8 | Helped |
| Use VGGNet instead of AlexNet for detection adaptation (ablation) | OM + MIL + FT · PASCAL VOC 2007 trainval | CorLoc (average): 49.8 -> 52.4 | Helped |
| Add detection fine-tuning to the OM + MIL pipeline (ablation) | Progressive domain adaptation with AlexNet · PASCAL VOC 2007 test | mAP: 23.4 -> 31.0 | Helped |
| Add class-specific proposal mining to MIL-based detector training (ablation) | MIL + FT-AlexNet · PASCAL VOC 2007 test | mAP: 23.0 -> 31.0 | Helped |
| Add MIL mining before detector fine-tuning (ablation) | OM + FT-AlexNet · PASCAL VOC 2007 test | mAP: 19.5 -> 31.0 | Helped |
| Use VGGNet instead of AlexNet for the full pipeline (ablation) | OM + MIL + FT · PASCAL VOC 2007 test | mAP: 31.0 -> 39.5 | Helped |
| Use the In-Out contrast score instead of Whole-Out (ablation) | Mask-out proposal ranking · PASCAL VOC 2007 trainval | CorLoc (top 50 proposals): 76.0 -> 82.9 | Helped |
| Use the In-Out contrast score instead of region score alone (ablation) | Mask-out proposal ranking · PASCAL VOC 2007 trainval | CorLoc: 32.7 -> 31.8 (top 1); 71.0 -> 73.8 (top 10) | Mixed |
| Train the target detector on image-level labels alone, without instance-level annotations | Progressive domain adaptation · PASCAL VOC 2007 test | mAP: +8 points | Helped |
Closest papers by meaning
- Weakly Supervised Object Localization As Domain Adaption (CVPR 2022 · similarity 0.90)
- Cross-Domain Weakly-Supervised Object Detection Through Progressive Domain Adaptation (CVPR 2018 · similarity 0.90)
- Rethinking the Route Towards Weakly Supervised Object Localization (CVPR 2020 · similarity 0.89)
- Bridging the Gap Between Classification and Localization for Weakly Supervised Object Localization (CVPR 2022 · similarity 0.89)
- Learning to Localize Objects with Noisy Labeled Instances (AAAI 2019 · similarity 0.89)
- You Reap What You Sow: Using Videos to Generate High Precision Object Proposals for Weakly-Supervised Object Detection (CVPR 2019 · 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.