Research map / Detection and segmentation / Weakly supervised detection and segmentation
Learning to Localize Objects with Noisy Labeled Instances
Read the paperCheck 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 multi-context features to WSDDN (ablation) | WSDDN · PASCAL VOC 2007 trainval | CorLoc (%): 55.1 -> 58.6 | Helped |
| add multi-context features to Fast RCNN refinement (ablation) | WSDDN + Fast RCNN · PASCAL VOC 2007 trainval | CorLoc (%): 61.7 -> 62.9 | Helped |
| add adaptive sampling (LSS-based latent update and reweighting) (ablation) | SD-LocNet · PASCAL VOC 2007 trainval | CorLoc (%): 62.9 -> 65.7 | Helped |
| add reliability-induced prior propagation (ablation) | SD-LocNet · PASCAL VOC 2007 trainval | CorLoc (%): 65.7 -> 67.4 | Helped |
| propagate priors from all mined instances without reliability selection (ablation) | SD-LocNet · PASCAL VOC 2007 trainval | CorLoc (%): 65.7 -> 63.1 | Hurt |
| increase clusters K from 1 to 3 (ablation) | SD-LocNet · PASCAL VOC 2007 trainval | CorLoc (%): around +1% | Helped |
| retain only top Rp=100 candidate positives (ablation) | SD-LocNet · PASCAL VOC 2007 trainval | CorLoc (%) | No clear effect |
| switch backbone from VGG-CNN-M to VGG-VD | SD-LocNet · PASCAL VOC 2007 trainval | CorLoc (%): 67.4 -> 70.9 | Helped |
Closest papers by meaning
- Rethinking the Route Towards Weakly Supervised Object Localization (CVPR 2020 · similarity 0.91)
- Evaluating Weakly Supervised Object Localization Methods Right (CVPR 2020 · similarity 0.90)
- Unveiling the Potential of Structure Preserving for Weakly Supervised Object Localization (CVPR 2021 · similarity 0.90)
- Weakly Supervised Object Localization As Domain Adaption (CVPR 2022 · similarity 0.90)
- Shallow Feature Matters for Weakly Supervised Object Localization (CVPR 2021 · similarity 0.89)
- Strengthen Learning Tolerance for Weakly Supervised Object Localization (CVPR 2021 · 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.