Research map / 3D vision / 3D generation
Modeling 3D Layout for Group Re-Identification
Quan Zhang, Kaiheng Dang, Jian-Huang Lai, Zhanxiang Feng, Xiaohua Xie
Check 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 3D layout modeling to appearance features (ablation) | 3DT · RoadGroup | Rank1: 88.89 -> 91.36 | Improved |
| Add 3D layout modeling to appearance features (ablation) | 3DT · RoadGroup | mAP: 91.25 -> 94.27 | Improved |
| Add depth to the 2D X-Y layout (ablation) | 3DT · RoadGroup | Rank1: 90.12 -> 91.36 | Improved |
| Add depth to the 2D X-Y layout (ablation) | 3DT · RoadGroup | mAP: 92.88 -> 94.27 | Improved |
| Use depth position alone instead of the full X-Y-D layout (ablation) | 3DT · RoadGroup | Rank1: 91.36 -> 87.65 | Hurt |
| Pretrain on City1M before finetuning | 3DT · CSG | Rank1: 92.9 -> 95.1 | Improved |
| Pretrain on City1M before finetuning | 3DT · DukeGroup | Rank1: 83.0 -> 90.9 | Improved |
| Pretrain on City1M before finetuning | 3DT · RoadGroup | Rank1: 91.4 -> 93.8 | Improved |
| Set layout sampling granularity σ to 2 rather than 10 (ablation) | 3DT · RoadGroup | mAP: 94.27 -> 92.92 | Hurt |
| Set layout sampling granularity σ to 50 rather than 10 (ablation) | 3DT · RoadGroup | Rank1: 91.36 -> 90.12 | Hurt |
| Set layout sampling granularity σ to 50 rather than 10 (ablation) | 3DT · RoadGroup | mAP: 94.27 -> 92.62 | Hurt |
Closest papers by meaning
- Separable Spatial-Temporal Residual Graph for Cloth-Changing Group Re-Identification (TPAMI 2024 · similarity 0.87)
- Learning Multi-Attention Context Graph for Group-Based Re-Identification (TPAMI 2020 · similarity 0.87)
- ID-Patch: Robust ID Association for Group Photo Personalization (CVPR 2025 · similarity 0.85)
- Unsupervised Group Re-identification via Adaptive Clustering-Driven Progressive Learning (AAAI 2024 · similarity 0.83)
- Alice Benchmarks: Connecting Real World Re-Identification with the Synthetic (ICLR 2024 · similarity 0.82)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 2025 · similarity 0.82)
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.
