Research map / Label-efficient and robust learning / Recognition and long-tail
Heterogeneous Relational Complement for Vehicle Re-Identification
Jiajian Zhao, Yifan Zhao, Jia Li, Ke Yan, Yonghong Tian
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 cross-level branch to the regional-only model (ablation) | HRCN · VeRi-776 | mAP: 0.807 -> 0.831 | Improved |
| add regional branch to the cross-level-only model (ablation) | HRCN · VeRi-776 | mAP: 0.814 -> 0.831 | Improved |
| use graph relation module instead of concatenation (ablation) | HRCN · VeRi-776 | mAP: 0.803 -> 0.831 | Improved |
| fuse bottom-top rather than into one unified graph (ablation) | HRCN · VeRi-776 | mAP: 0.803 -> 0.831 | Improved |
| add stage 3 features as lower-level complement (ablation) | HRCN · VeRi-776 | mAP: 0.819 -> 0.825 | Improved |
| add stage 2 features as well as stage 3 (ablation) | HRCN · VeRi-776 | mAP: 0.825 -> 0.831 | Improved |
| progressive center pooling instead of grid partition (ablation) | HRCN · VeRi-776 | mAP: 0.817 -> 0.831 | Improved |
| progressive center pooling instead of vertical partition (ablation) | HRCN · VeRi-776 | mAP: 0.818 -> 0.831 | Improved |
| progressive center pooling instead of horizontal partition (ablation) | HRCN · VeRi-776 | mAP: 0.822 -> 0.831 | Improved |
| use full HRCN instead of plain baseline | ResNet-50-ibn-a baseline · VeRi-776 | mAP: 0.768 -> 0.831 | Improved |
| HRCN versus PVEN with extra annotations | HRCN · VERI-Wild Small | CMC@1: 0.967 -> 0.940 | Hurt |
| HRCN versus PVEN on mCGM benchmark | HRCN · VeRi-776 benchmark | mCGM: 0.560 -> 0.630 | Improved |
Closest papers by meaning
- Infer the Whole from a Glimpse of a Part: Keypoint-Based Knowledge Graph for Vehicle Re-Identification (AAAI 2025 · similarity 0.88)
- Parsing-Based View-Aware Embedding Network for Vehicle Re-Identification (CVPR 2020 · similarity 0.87)
- Learning Coarse-to-Fine Structured Feature Embedding for Vehicle Re-Identification (AAAI 2018 · similarity 0.87)
- Viewpoint-Aware Attentive Multi-View Inference for Vehicle Re-Identification (CVPR 2018 · similarity 0.86)
- CLIP-driven View-aware Prompt Learning for Unsupervised Vehicle Re-identification (AAAI 2025 · similarity 0.85)
- Multi-Modal Object Re-identification via Sparse Mixture-of-Experts (ICML 2025 · similarity 0.84)
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.
