Research map / Detection and segmentation / Weakly supervised detection and segmentation
Self-Supervised Object Localization with Joint Graph Partition
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 joint graph partition over paired transformed images (ablation) | self-supervised localization baseline · CUB-200-2011 | GT-Known Loc: 54.6 -> 57.9 | Helped |
| add equivariant regularization loss LER (ablation) | joint-graph-partition localization model · CUB-200-2011 | GT-Known Loc: 57.9 -> 60.4 | Helped |
| add flipping to rotation transformations (ablation) | self-supervised localization model · CUB-200-2011 | GT-Known Loc: 57.9 -> 58.1 | Helped |
| add rescaling to rotation transformations (ablation) | self-supervised localization model · CUB-200-2011 | GT-Known Loc: 57.9 -> 58.6 | Helped |
| add translation to rotation transformations (ablation) | self-supervised localization model · CUB-200-2011 | GT-Known Loc: 57.9 -> 58.2 | Helped |
| add flipping and rescaling to rotation transformations (ablation) | self-supervised localization model · CUB-200-2011 | GT-Known Loc: 57.9 -> 60.2 | Helped |
| add flipping and translation to rotation transformations (ablation) | self-supervised localization model · CUB-200-2011 | GT-Known Loc: 57.9 -> 60.0 | Helped |
| add rescaling and translation to rotation transformations (ablation) | self-supervised localization model · CUB-200-2011 | GT-Known Loc: 57.9 -> 60.2 | Helped |
| use all four transformation types: rotation, flipping, rescaling, and translation (ablation) | self-supervised localization model · CUB-200-2011 | GT-Known Loc: 57.9 -> 60.4 | Helped |
| replace feature addition with multiplication (ablation) | joint-graph-partition localization model · CUB-200-2011 | GT-Known Loc: 60.4 -> 57.8 | Hurt |
| replace feature addition with concatenation (ablation) | joint-graph-partition localization model · CUB-200-2011 | GT-Known Loc: 60.4 -> 59.2 | Hurt |
| use the proposed self-supervised localization method instead of PSY | VGG16 localization models · CUB-200-2011 | GT-Known Loc: 83.78 -> 88.83 | Helped |
Closest papers by meaning
- PsyNet: Self-Supervised Approach to Object Localization Using Point Symmetric Transformation (AAAI 2020 · similarity 0.91)
- Unveiling the Potential of Structure Preserving for Weakly Supervised Object Localization (CVPR 2021 · similarity 0.90)
- Rethinking the Route Towards Weakly Supervised Object Localization (CVPR 2020 · similarity 0.90)
- Bridging the Gap Between Classification and Localization for Weakly Supervised Object Localization (CVPR 2022 · similarity 0.89)
- Evaluating Weakly Supervised Object Localization Methods Right (CVPR 2020 · similarity 0.89)
- MutualVPR: A Mutual Learning Framework for Resolving Supervision Inconsistencies via Adaptive Clustering (NeurIPS 2025 · 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.