Research map / 3D vision / Face and shape
Learning 3D Shape Feature for Texture-Insensitive Person Re-Identification
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 3D shape learning with ASSP and MGS to the two-branch baseline (ablation) | 3DSL person ReID model · PRCC | rank-1: 45.6 -> 51.3 | Helped |
| add 3D shape learning with ASSP and MGS to the two-branch baseline (ablation) | 3DSL person ReID model · Market1501 | mAP: 84.8 -> 87.3 | Helped |
| use 3D shape features instead of RGB features alone (ablation) | 3DSL person ReID model · PRCC | rank-1: 45.6 -> 50.4 | Helped |
| use texture-informed triplet sampling rather than random triplet sampling (ablation) | RGB feature branch · PRCC | rank-1: 50.4 -> 51.3 | Helped |
| add MGS global and part shape features (ablation) | 3DSL · PRCC | rank-1: 50.1 -> 51.3 | Helped |
| add MGS global and part shape features (ablation) | 3DSL · Market1501 | mAP: 86.7 -> 87.3 | Helped |
| add self-supervised keypoint and silhouette projection losses (ablation) | 3DSL with adversarial loss · PRCC | rank-1: 47.0 -> 51.3 | Helped |
| add adversarial learning to keypoint and silhouette projection (ablation) | 3DSL with MGS · PRCC | rank-1: 49.8 -> 51.3 | Helped |
| add keypoint projection loss Lkey (ablation) | 3DSL with MGS and the other ASSP losses · PRCC | rank-1: 49.5 -> 51.3 | Helped |
| add silhouette projection loss Lsil (ablation) | 3DSL with MGS and the other ASSP losses · PRCC | rank-1: 49.0 -> 51.3 | Helped |
| use the full model rather than the strongest reported comparison | 3DSL · FGPR (All groups) | rank-1: 87.1 -> 88.0 | Helped |
| use the full model rather than the strongest reported comparison | 3DSL · FGPR (Blue group) | rank-1: 93.6 -> 88.3 | Hurt |
Closest papers by meaning
- Distilling CLIP with Dual Guidance for Learning Discriminative Human Body Shape Representation (CVPR 2024 · similarity 0.90)
- Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions (CVPR 2024 · similarity 0.87)
- DIFFER: Disentangling Identity Features via Semantic Cues for Clothes-Changing Person Re-ID (CVPR 2025 · similarity 0.86)
- From Poses to Identity: Training-Free Person Re-Identification via Feature Centralization (CVPR 2025 · similarity 0.86)
- Exploring the Camera Bias of Person Re-identification (ICLR 2025 · similarity 0.86)
- SEAS: ShapE-Aligned Supervision for Person Re-Identification (CVPR 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.