Research map / Label-efficient and robust learning / Recognition and long-tail
Pose Transferrable Person Re-Identification
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 labeled pose-transferred samples generated from MARS | ResNet-50 ReID model · Market1501 | rank-1: 73.90 -> 79.75 | Helped |
| Add labeled pose-transferred samples generated from MARS | ResNet-50 ReID model · DukeMTMC-reID | rank-1: 65.22 -> 68.64 | Helped |
| Add labeled pose-transferred samples generated from MARS | ResNet-50 ReID model · CUHK03 labeled | rank-1: 22.2 -> 33.8 | Helped |
| Add labeled pose-transferred samples instead of unlabeled GAN samples | ResNet-50 ReID model · Market1501 | rank-1: 78.06 -> 79.75 | Helped |
| Add labeled pose-transferred samples instead of unlabeled GAN samples | ResNet-50 ReID model · DukeMTMC-reID | rank-1: 67.68 -> 68.64 | Helped |
| Add a ReID guider during pose-transfer generation (ablation) | Pose-transfer augmentation · Market1501 | rank-1: 76.93 -> 79.75 | Helped |
| Add a ReID guider during pose-transfer generation (ablation) | Pose-transfer augmentation · DukeMTMC-reID | rank-1: 66.70 -> 68.64 | Helped |
| Add a ReID guider during pose-transfer generation (ablation) | Pose-transfer augmentation · CUHK03 labeled | rank-1: 28.1 -> 33.8 | Helped |
| Add a ReID guider during pose-transfer generation (ablation) | Pose-transfer augmentation · CUHK03 detected | rank-1: 25.2 -> 30.1 | Helped |
| Use two pose-transferred samples per target image rather than one (ablation) | ResNet-50 ReID model · Market1501, DukeMTMC-reID, and CUHK03 | rank-1 | Helped |
| Add still more pose-transferred samples beyond two per target image (ablation) | ResNet-50 ReID model · Market1501, DukeMTMC-reID, and CUHK03 | rank-1 | Hurt |
| Use label smoothing for generated samples, with ε set to 0.4 (ablation) | Cross-entropy ReID training · CUHK03 | rank-1 | Helped |
Closest papers by meaning
- Adversarially Occluded Samples for Person Re-Identification (CVPR 2018 · similarity 0.90)
- Learning Disentangled Representation for Robust Person Re-identification (NeurIPS 2019 · similarity 0.90)
- From Poses to Identity: Training-Free Person Re-Identification via Feature Centralization (CVPR 2025 · similarity 0.90)
- Exploring the Camera Bias of Person Re-identification (ICLR 2025 · similarity 0.88)
- Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions (CVPR 2024 · similarity 0.87)
- Person De-reidentification: A Variation-guided Identity Shift Modeling (CVPR 2025 · 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.