Research map / Video / Video segmentation and tracking
Temporal Knowledge Propagation for Image-to-Video Person Re-Identification
Xinqian Gu, Bingpeng Ma, Hong Chang, Shiguang Shan, Xilin Chen
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 TKP-F feature propagation (ablation) | image-to-video Re-ID baseline · MARS I2V | mAP: 55.5 -> 64.2 | Improved |
| add TKP-F feature propagation (ablation) | baseline · Duke I2V | mAP: 65.6 -> 74.2 | Improved |
| add TKP-D distance propagation (ablation) | baseline · MARS I2V | mAP: 55.5 -> 63.1 | Improved |
| add TKP-D distance propagation (ablation) | baseline · Duke I2V | mAP: 65.6 -> 74.9 | Improved |
| combine TKP-F and TKP-D (ablation) | TKP-F · MARS I2V | top-1 accuracy: 75.0 -> 75.6 | Improved |
| add TKP (ablation) | baseline image network · MARS I2I | mAP: 49.2 -> 55.0 | Improved |
| add TKP (ablation) | baseline video network · MARS V2V | mAP: 72.6 -> 73.3 | No clear effect |
| add non-local blocks (ablation) | TKP-F · MARS I2V | mAP: 57.8 -> 64.2 | Improved |
| block TKP gradient into video network (ablation) | TKP-F · MARS V2V | top-1 accuracy: 78.5 -> 83.2 | Improved |
| block TKP gradient into video network (ablation) | TKP-D · MARS I2V | top-1 accuracy: 74.2 -> 75.0 | Improved |
| learn video and image networks jointly instead of pre-trained video model (ablation) | TKP · MARS I2V | top-1 accuracy: 73.2 -> 75.6 | Improved |
| add classification loss to integrated triplet loss (ablation) | baseline · MARS I2V | top-1 accuracy: 59.1 -> 67.1 | Improved |
Closest papers by meaning
- Relation-Guided Spatial Attention and Temporal Refinement for Video-Based Person Re-Identification (AAAI 2020 · similarity 0.88)
- Rethinking Temporal Fusion for Video-Based Person Re-Identification on Semantic and Time Aspect (AAAI 2020 · similarity 0.88)
- Temporal-Enhanced Convolutional Network for Person Re-Identification (AAAI 2018 · similarity 0.88)
- READ: Reciprocal Attention Discriminator for Image-to-Video Re-Identification (ECCV 2020 · similarity 0.88)
- Video-Based Person Re-Identification With Spatial and Temporal Memory Networks (ICCV 2021 · similarity 0.88)
- Temporal Correlation Vision Transformer for Video Person Re-Identification (AAAI 2024 · similarity 0.87)
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.
