Research map / Label-efficient and robust learning / Recognition and long-tail
Hierarchical Hyperbolic Representation Learning for Aerial-Ground Person Re-Identification
QiWei Yang
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 TMF (text-guided multi-granularity fusion) (ablation) | CLIP baseline · CARGO A->G | mAP: 66.06 -> 69.50 | Improved |
| add TMF (ablation) | CLIP baseline · CARGO ALL | mAP: 63.61 -> 65.46 | Improved |
| add HHL on top of TMF (ablation) | Baseline+TMF · CARGO ALL | mAP: 65.46 -> 67.52 | Improved |
| add prototype-to-child entailment loss (ablation) | HiHR · CARGO ALL | mAP / Rank-1: 64.62/71.15 to 65.46/72.12 | Improved |
| add visual-prompt consistency entailment terms (ablation) | HiHR · CARGO ALL | mAP: 65.71 -> 67.52 | Improved |
| use dual-level prompts instead of text-only (ablation) | HiHR · CARGO ALL | mAP: 64.58 -> 67.52 | Improved |
| use dual-level prompts instead of CoCoOp (ablation) | HiHR · CARGO ALL | mAP: 66.73 -> 67.52 | Improved |
| use TMF instead of standard cross-attention fusion (ablation) | HiHR · CARGO ALL | mAP / Rank-1: +1.72 mAP and +3.21 Rank-1 | Improved |
| use cross-attention instead of mean fusion (ablation) | HiHR · CARGO ALL | mAP: 64.95 -> 65.80 | Improved |
| use layers {5,8,12} instead of final layer only or adjacent {10,11,12} (ablation) | HiHR · CARGO | mAP | Improved |
| vary entailment weight lambda_e (ablation) | HiHR · CARGO | mAP | No clear effect |
| use small curvature initialization instead of 1.0-4.0 (ablation) | HiHR · CARGO | mAP | Hurt |
Closest papers by meaning
- Bridging the Sky and Ground: Towards View-Invariant Feature Learning for Aerial-Ground Person Re-Identification (ICCV 2025 · similarity 0.92)
- AG-VPReID: A Challenging Large-Scale Benchmark for Aerial-Ground Video-based Person Re-Identification (CVPR 2025 · similarity 0.89)
- View-decoupled Transformer for Person Re-identification under Aerial-ground Camera Network (CVPR 2024 · similarity 0.89)
- GSAlign: Geometric and Semantic Alignment Network for Aerial-Ground Person Re-Identification (NeurIPS 2025 · similarity 0.88)
- SeCap: Self-Calibrating and Adaptive Prompts for Cross-view Person Re-Identification in Aerial-Ground Networks (CVPR 2025 · similarity 0.87)
- High-Order Structure Based Middle-Feature Learning for Visible-Infrared Person Re-identification (AAAI 2024 · similarity 0.86)
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.
