Research map / Detection and segmentation / CNN features
Detangling People: Individuating Multiple Close People and Their Body Parts via Region Assembly
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 |
|---|---|---|---|
| Use global region assembly instead of connected-component grouping | Person instance segmentation · UCI | instance mean IoU (F): 41.62 -> 63.02 | Helped |
| Use global region assembly instead of greedy sequential grouping | Person instance segmentation · UCI | instance mean IoU (F): 46.88 -> 63.02 | Helped |
| Use region assembly rather than the strongest reported baseline | Person instance segmentation · UCI | instance mean IoU (F): 61.81 -> 63.02 | Helped |
| Use region assembly rather than the strongest reported baseline | Person instance segmentation · MPII | instance mean IoU (F): 56.04 -> 57.48 | Helped |
| Use region assembly rather than CNN pose detection | Person instance segmentation · MPII | instance mean IoU (B): 45.48 -> 57.15 | Helped |
| Use region assembly rather than the strongest reported baseline | Body part segmentation · UCI | part mean IoU (F): 37.98 -> 38.39 | Helped |
| Use region assembly rather than greedy grouping | Body part segmentation · UCI | part mean IoU (B): 32.30 -> 38.56 | Helped |
| Use region assembly rather than NBest pose masks | Body part segmentation · MPII | part mean IoU (F): 28.71 -> 35.48 | Helped |
| Use region assembly rather than NBest pose masks | Body part segmentation · MPII | part mean IoU (B): 29.16 -> 35.47 | Helped |
| Apply region assembly without training on Buffy | Person individuation · Buffy episode 4, 5, 6 | forward score: 62.4% -> 68.22% | Helped |
| Replace direct whole-person detection with assembling candidate body-part regions into instances | Multi-person instance and part segmentation · UCI and MPII | instance mean IoU | Helped |
Closest papers by meaning
- DeepCut: Joint Subset Partition and Labeling for Multi Person Pose Estimation (CVPR 2016 · similarity 0.91)
- ArtTrack: Articulated Multi-Person Tracking in the Wild (CVPR 2017 · similarity 0.89)
- Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields (CVPR 2017 · similarity 0.89)
- GigaHumanDet: Exploring Full-Body Detection on Gigapixel-Level Images (AAAI 2024 · similarity 0.87)
- PairDETR : Joint Detection and Association of Human Bodies and Faces (CVPR 2024 · similarity 0.87)
- PBADet: A One-Stage Anchor-Free Approach for Part-Body Association (ICLR 2024 · 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.