Research map / Label-efficient and robust learning / Domain adaptation and generalization
Large-Scale Training Data Search for Object 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 |
|---|---|---|---|
| Use target-specific subset search without budget pruning instead of the full source pool | IDE re-ID training · AlicePerson | rank-1 accuracy: 38.96 -> 48.19 | Helped |
| Use target-specific subset search without budget pruning instead of the full source pool | IDE re-ID training · AliceVehicle | rank-1 accuracy: 30.47 -> 46.78 | Helped |
| Use SnP rather than random sampling at the 2% ID budget | IDE re-ID training · AlicePerson | rank-1 accuracy: +14.56% | Helped |
| Use SnP rather than greedy sampling at the 2% ID budget | IDE re-ID training · AlicePerson | rank-1 accuracy: +5.01% | Helped |
| Use SnP rather than random or greedy sampling at the 2% ID budget | IDE re-ID training · AliceVehicle | rank-1 accuracy: +7.13%; +6.66% | Helped |
| Prune the searched set to 2% of source IDs (ablation) | SnP · AlicePerson | rank-1 accuracy: -9.23% | Hurt |
| Prune the searched set to 40% of images (ablation) | SnP · AlicePerson | rank-1 accuracy: -9.96% | Hurt |
| Use furthest point sampling rather than random image sampling (ablation) | SnP image pruning · Four targets | rank-1 accuracy | Helped |
| Increase the number of clusters J (ablation) | SnP target-specific search · AlicePerson | rank-1 accuracy | Helped |
| Add synthetic datasets to the source pool (ablation) | SnP · AlicePerson | rank-1 accuracy: 14.17 -> 25.44 -> 36.32 -> 38.23 | Helped |
| Train on the AlicePerson-specific searched set rather than the source pool | IDE re-ID training · MSMT | rank-1 accuracy: 35.72 -> 8.19 | Hurt |
Closest papers by meaning
- Alice Benchmarks: Connecting Real World Re-Identification with the Synthetic (ICLR 2024 · similarity 0.91)
- Uncertainty-Aware Multi-Shot Knowledge Distillation for Image-Based Object Re-Identification (AAAI 2020 · similarity 0.89)
- Asymmetric Co-Teaching for Unsupervised Cross-Domain Person Re-Identification (AAAI 2020 · similarity 0.88)
- Camera-Aware Proxies for Unsupervised Person Re-Identification (AAAI 2021 · similarity 0.87)
- Cheb-GR: Rethinking K-nearest Neighbor Search in Re-ranking for Person Re-identification (CVPR 2025 · similarity 0.85)
- Catalyst for Clustering-Based Unsupervised Object Re-identification: Feature Calibration (AAAI 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.