Research map / Video / Video segmentation and tracking
Temporal-Enhanced Convolutional Network for Person Re-Identification
Read the paperCheck 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 temporal convolution network to the RCN (color + optical flow) (ablation) | RCN · PRID 2011 | CMC rank-1 accuracy (%): 77.4 -> 81.1 | Helped |
| add temporal convolution network to the RCN (color + optical flow) (ablation) | RCN · iLIDS-VID | CMC rank-1 accuracy (%): 51.3 -> 60.6 | Helped |
| add temporal convolution network with color-only input (ablation) | RCN color only · iLIDS-VID | CMC rank-1 accuracy (%): 38.9 -> 48.0 | Helped |
| add temporal convolution network with color-only input (ablation) | RCN color only · PRID 2011 | CMC rank-1 accuracy (%): 74.7 -> 76.7 | Helped |
| add optical flow input to T-CN (ablation) | T-CN · iLIDS-VID | CMC rank-1 accuracy (%): 48.0 -> 60.6 | Helped |
| learned weighting (full-length 1D conv) instead of average pooling for kernel responses (ablation) | T-CN · PRID 2011 | CMC rank-1 accuracy (%): 75.1 -> 81.1 | Helped |
| learned weighting (full-length 1D conv) instead of average pooling for kernel responses (ablation) | T-CN · iLIDS-VID | CMC rank-1 accuracy (%): 57.5 -> 60.6 | Helped |
| increase temporal kernel size from 2 to 4 (ablation) | T-CN · PRID 2011 | CMC rank-1 accuracy (%): 81.1 -> 82.0 | No clear effect |
| increase temporal kernel size from 2 to 6 or 8 (ablation) | T-CN · PRID 2011 | CMC rank-1 accuracy (%): 81.1 -> 77.8 (S=6), 80.0 (S=8) | Hurt |
| T-CN versus RCN in cross-dataset testing | T-CN · Cross-dataset PRID 2011 / iLIDS-VID | CMC rank-1 accuracy (%): 24.0 vs 25.0 (PRID), 13.4 vs 11.0 (iLIDS-VID) | Mixed |
Closest papers by meaning
- Recurrent Convolutional Network for Video-Based Person Re-Identification (CVPR 2016 · similarity 0.92)
- Spatial-Temporal Graph Convolutional Network for Video-Based Person Re-Identification (CVPR 2020 · similarity 0.91)
- See the Forest for the Trees: Joint Spatial and Temporal Recurrent Neural Networks for Video-Based Person Re-Identification (CVPR 2017 · similarity 0.91)
- Multi-Scale 3D Convolution Network for Video Based Person Re-Identification (AAAI 2019 · similarity 0.90)
- BiCnet-TKS: Learning Efficient Spatial-Temporal Representation for Video Person Re-Identification (CVPR 2021 · similarity 0.89)
- Multi-Rate Gated Recurrent Convolutional Networks for Video-Based Pedestrian Re-Identification (AAAI 2018 · similarity 0.89)
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.