Research map / Detection and segmentation / CNN features
Part-Regularized Near-Duplicate Vehicle Re-Identification
Check 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 |
|---|---|---|---|
| add part-regularized local branches for window, lights, and brand (ablation) | ResNet-50 vehicle Re-ID · VehicleID | CMC@1: 0.645 -> 0.742 | Helped |
| add part-regularized local branches for window, lights, and brand (ablation) | ResNet-50 vehicle Re-ID · VehicleID | CMC@5: 0.800 -> 0.864 | Helped |
| add the window branch (ablation) | Global+Light+Brand vehicle Re-ID · VehicleID | CMC@1: 0.675 -> 0.742 | Helped |
| add the brand branch (ablation) | Global+Light+Window vehicle Re-ID · VehicleID | CMC@1: 0.710 -> 0.742 | Helped |
| add the light branch (ablation) | Global+Window+Brand vehicle Re-ID · VehicleID | CMC@1: 0.726 -> 0.742 | Helped |
| add the global branch (ablation) | Window+Light+Brand vehicle Re-ID · VehicleID | CMC@1: 0.687 -> 0.742 | Helped |
| increase input resolution from 128 × 128 to 256 × 256 (ablation) | ResNet-50 vehicle Re-ID · VehicleID small, medium, and large | CMC@1: 0.726 -> 0.784; 0.685 -> 0.750; 0.661 -> 0.742 | Helped |
| increase input resolution from 128 × 128 to 512 × 512 (ablation) | ResNet-50 vehicle Re-ID · VeRi-776 | mAP: 0.653 -> 0.743 | Helped |
| increase input resolution from 128 × 128 to 512 × 512 (ablation) | ResNet-50 vehicle Re-ID · VeRi-776 | HIT@1: 0.878 -> 0.943 | Helped |
| use the part-regularized model rather than VAMI | Vehicle Re-ID · VehicleID small, medium, and large | CMC@1: 0.631 -> 0.784; 0.529 -> 0.750; 0.473 -> 0.742 | Helped |
| use the part-regularized model rather than Siamese+Path | Vehicle Re-ID · VeRi-776 | mAP: 0.583 -> 0.743 | Helped |
Closest papers by meaning
- Learning Coarse-to-Fine Structured Feature Embedding for Vehicle Re-Identification (AAAI 2018 · similarity 0.90)
- Deep Relative Distance Learning: Tell the Difference Between Similar Vehicles (CVPR 2016 · similarity 0.88)
- Viewpoint-Aware Attentive Multi-View Inference for Vehicle Re-Identification (CVPR 2018 · similarity 0.87)
- Infer the Whole from a Glimpse of a Part: Keypoint-Based Knowledge Graph for Vehicle Re-Identification (AAAI 2025 · similarity 0.87)
- CLIP-driven View-aware Prompt Learning for Unsupervised Vehicle Re-identification (AAAI 2025 · similarity 0.85)
- Alice Benchmarks: Connecting Real World Re-Identification with the Synthetic (ICLR 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.