Research map / Label-efficient and robust learning / Domain adaptation and generalization
A Novel Unsupervised Camera-Aware Domain Adaptation Framework for Person Re-Identification
Lei Qi, Lei Wang, Jing Huo, Luping Zhou, Yinghuan Shi, Yang Gao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 unsupervised online triplet generation (UOT) (ablation) | BL · Duke->Market1501 | mAP: 19.4 -> 27.4 | Improved |
| add unsupervised online triplet generation (UOT) (ablation) | BL · Market1501->Duke | mAP: 21.3 -> 27.5 | Improved |
| compute triplet distances with re-ranking instead of Euclidean (ablation) | BL+UOT · Duke->Market1501 | mAP: 23.6 -> 27.4 | Improved |
| add camera-aware adversarial learning with CCE (ablation) | BL+UOT · Duke->Market1501 | mAP: 27.4 -> 34.5 | Improved |
| add camera-aware adversarial learning with CCE (ablation) | BL+UOT · Market1501->Duke | mAP: 27.5 -> 36.7 | Improved |
| use CCE instead of GRL in camera-aware adversarial learning (ablation) | BL+UOT+CAL · Market1501->Duke | Rank-1: 47.7 -> 55.4 | Improved |
| use camera-aware instead of domain-aware adversarial learning (ablation) | BL+UOT · Duke->Market1501 | mAP: 25.5 -> 34.5 | Improved |
| add CAL-CCE without UOT (ablation) | BL · Duke->Market1501 | mAP: 19.4 -> 8.4 | Hurt |
| add CAL-CCE without UOT (ablation) | BL · Market1501->Duke | mAP: 21.3 -> 23.8 | Improved |
| use CCE instead of all-others equiprobability (AOE) (ablation) | CAL · Duke->Market1501 | mAP: 29.6 -> 34.5 | Improved |
| generate triplets online instead of offline (ablation) | UOT · Duke->Market1501 | mAP: 13.5 -> 27.4 | Improved |
| add CAL-CCE to semi-supervised triplets (ablation) | BL+SOT · Duke->Market1501 | mAP: 39.2 -> 49.6 | Improved |
Closest papers by meaning
- Camera-Driven Representation Learning for Unsupervised Domain Adaptive Person Re-identification (ICCV 2023 · similarity 0.94)
- Unsupervised Multi-Source Domain Adaptation for Person Re-Identification (CVPR 2021 · similarity 0.93)
- Joint Disentangling and Adaptation for Cross-Domain Person Re-Identification (ECCV 2020 · similarity 0.93)
- Domain Adaptive Attention Learning for Unsupervised Person Re-Identification (AAAI 2020 · similarity 0.92)
- Towards Discriminative Representation Learning for Unsupervised Person Re-Identification (ICCV 2021 · similarity 0.92)
- Learning to Adapt Across Dual Discrepancy for Cross-Domain Person Re-Identification (TPAMI 2022 · similarity 0.92)
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.
