Research map / Detection and segmentation / Open-set and few-shot detection
Unsupervised Part Segmentation Through Disentangling Appearance and Shape
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 part classification loss (ablation) | Full model · CelebA wild, K=8 | landmark regression error: 14.38 -> 12.26 | Helped |
| add reconstruction loss (ablation) | Full model · CelebA wild, K=8 | landmark regression error: 46.01 -> 12.26 | Helped |
| add foreground concentration loss (ablation) | Full model · CelebA wild, K=8 | landmark regression error: 41.42 -> 12.26 | Helped |
| add background concentration loss (ablation) | Full model · CelebA wild, K=8 | landmark regression error: 29.93 -> 12.26 | Helped |
| use the proposed method instead of SCOPS | CUB bird categories · CUB-1, CUB-2, and CUB-3 | landmark regression mean error: 18.50 -> 18.15; 18.82 -> 17.54; 21.07 -> 19.40 | Helped |
| use the proposed method instead of SCOPS | PASCAL VOC categories · PASCAL VOC sheep, bus, motor, cow, and car | IoU: 56.95 -> 64.87; 73.82 -> 78.17; 58.53 -> 59.98; 60.79 -> 65.60; 65.18 -> 70.49 | Helped |
| use the proposed method instead of SCOPS | PASCAL VOC categories · PASCAL VOC horse and areo | IoU: 55.76 -> 53.65; 69.02 -> 52.82 | Mixed |
| use the proposed method instead of IMM | CelebA wild · CelebA wild, K=8 | landmark regression error: 8.74 -> 12.26 | Hurt |
Closest papers by meaning
- Unsupervised Part-Based Disentangling of Object Shape and Appearance (CVPR 2019 · similarity 0.92)
- Unsupervised Part Discovery from Contrastive Reconstruction (NeurIPS 2021 · similarity 0.91)
- unMORE: Unsupervised Multi-Object Segmentation via Center-Boundary Reasoning (ICML 2025 · similarity 0.88)
- Segment Anything without Supervision (NeurIPS 2024 · similarity 0.87)
- Hierarchy-Agnostic Unsupervised Segmentation: Parsing Semantic Image Structure (NeurIPS 2024 · similarity 0.87)
- Multi-Part Object Representations via Graph Structures and Co-Part Discovery (NeurIPS 2026 · similarity 0.87)
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.