Research map / Video / Video segmentation and tracking
Learning Deep Neural Networks for Vehicle Re-ID With Visual-Spatio-Temporal Path Proposals
Yantao Shen, Tong Xiao, Hongsheng Li, Shuai Yi, Xiaogang Wang
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 Path-LSTM on proposed visual-spatio-temporal paths (ablation) | Siamese-CNN · VeRi-776 | mAP: 54.21 -> 58.27 | Improved |
| add Path-LSTM on proposed visual-spatio-temporal paths (ablation) | Siamese-CNN · VeRi-776 | top-1 accuracy: 79.32 -> 83.49 | Improved |
| add Siamese-CNN score to Path-LSTM (ablation) | Path-LSTM only · VeRi-776 | mAP: 54.49 -> 58.27 | Improved |
| add Siamese-CNN score to Path-LSTM (ablation) | Path-LSTM only · VeRi-776 | top-1 accuracy: 82.89 -> 83.49 | Improved |
| replace empirical average with trained Path-LSTM to score paths (ablation) | Chain MRF model · VeRi-776 | mAP: 44.31 -> 54.49 | Improved |
| replace empirical average with trained Path-LSTM to score paths (ablation) | Chain MRF model · VeRi-776 | top-1 accuracy: 54.41 -> 82.89 | Improved |
| add spatio-temporal branch to Siamese-Visual (ablation) | Siamese-Visual · VeRi-776 | mAP: 29.48 -> 54.21 | Improved |
| add spatio-temporal branch to Siamese-Visual (ablation) | Siamese-Visual · VeRi-776 | top-1 accuracy: 41.12 -> 79.32 | Improved |
| learn spatio-temporal relation with a network instead of hand-set STR score (ablation) | Siamese-Visual+STR · VeRi-776 | mAP: 40.26 -> 54.21 | Improved |
| use ResNet50 backbone instead of VGG16 for full model (ablation) | Siamese-CNN+Path-LSTM · VeRi-776 | mAP: 46.85 -> 58.27 | Improved |
| add Path-LSTM on VGG16 Siamese-CNN (ablation) | Siamese-CNN-VGG16 · VeRi-776 | mAP: 44.32 -> 45.56 | Improved |
| add Path-LSTM and Siamese-CNN combination on VGG16 (ablation) | Siamese-CNN-VGG16 · VeRi-776 | mAP: 44.32 -> 46.85 | Improved |
Closest papers by meaning
- Deep Relative Distance Learning: Tell the Difference Between Similar Vehicles (CVPR 2016 · similarity 0.89)
- Part-Regularized Near-Duplicate Vehicle Re-Identification (CVPR 2019 · similarity 0.88)
- Viewpoint-Aware Attentive Multi-View Inference for Vehicle Re-Identification (CVPR 2018 · similarity 0.88)
- Learning Coarse-to-Fine Structured Feature Embedding for Vehicle Re-Identification (AAAI 2018 · similarity 0.87)
- Gated Siamese Convolutional Neural Network Architecture for Human Re-identification (ECCV 2016 · similarity 0.87)
- Infer the Whole from a Glimpse of a Part: Keypoint-Based Knowledge Graph for Vehicle Re-Identification (AAAI 2025 · 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.
