Research map / Label-efficient and robust learning / Domain adaptation and generalization
Joint Disentangling and Adaptation for Cross-Domain Person Re-Identification
Yang Zou, Xiaodong Yang, Zhiding Yu, B.V.K. Vijaya Kumar, Jan Kautz
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 the disentangling module (ablation) | source-only baseline · Market->Duke | mAP: 19.3 -> 24.2 | Improved |
| add the disentangling module (ablation) | source-only baseline · Duke->Market | mAP: 15.0 -> 26.8 | Improved |
| add disentangling to adversarial alignment (ablation) | +A · Market->Duke | mAP: +8.6% mAP | Improved |
| add disentangling to adversarial alignment (ablation) | +A · Duke->Market | mAP: +6.4% mAP | Improved |
| add disentangling to self-training (ablation) | +ST · Market->Duke | mAP: +4.0% mAP | Improved |
| add disentangling to self-training (ablation) | +ST · Duke->Market | mAP: +5.7% mAP | Improved |
| add adversarial alignment to disentangling + self-training (ablation) | +D+ST · Market->Duke | mAP: 58.4 -> 63.8 | Improved |
| add adversarial alignment to disentangling + self-training (ablation) | +D+ST · Duke->Market | mAP: 56.5 -> 61.7 | Improved |
| add self-training to disentangling + adversarial alignment (ablation) | +D+A · Market->Duke | mAP: 36.3 -> 63.8 | Improved |
| add self-training to disentangling + adversarial alignment (ablation) | +D+A · Duke->Market | mAP: 28.6 -> 61.7 | Improved |
| replace DG-Net self-training with joint disentangling and adaptation (DG-Net++) (ablation) | DG-Net + self-training · Market->Duke | mAP: 54.6 -> 63.8 | Improved |
| add target-domain self-training on within-domain DG-Net disentangling (ablation) | DG-Net · Market->Duke | mAP: 54.6 -> 52.9 | Hurt |
Closest papers by meaning
- Learning to Adapt Across Dual Discrepancy for Cross-Domain Person Re-Identification (TPAMI 2022 · similarity 0.93)
- Image-Image Domain Adaptation With Preserved Self-Similarity and Domain-Dissimilarity for Person Re-Identification (CVPR 2018 · similarity 0.93)
- Cross-Dataset Person Re-Identification via Unsupervised Pose Disentanglement and Adaptation (ICCV 2019 · similarity 0.93)
- A Novel Unsupervised Camera-Aware Domain Adaptation Framework for Person Re-Identification (ICCV 2019 · similarity 0.93)
- Towards Discriminative Representation Learning for Unsupervised Person Re-Identification (ICCV 2021 · similarity 0.92)
- Self-Similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-Identification (ICCV 2019 · 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.
