Research map / Video / Action recognition
3D Local Convolutional Neural Networks for Gait Recognition
Zhen Huang, Dixiu Xue, Xu Shen, Xinmei Tian, Houqiang Li, Jianqiang Huang, Xian-Sheng Hua
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 3D local block to Block 3 (ablation) | GaitPart-style backbone · CASIA-B CL (64x44) | rank-1 accuracy: 76.7 -> 81.6 | Improved |
| add 3D local block to Block 1 (ablation) | GaitPart-style backbone · CASIA-B CL (64x44) | rank-1 accuracy: 76.7 -> 77.7 | Improved |
| add 3D local block to Block 2 (ablation) | GaitPart-style backbone · CASIA-B CL (64x44) | rank-1 accuracy: 76.7 -> 79.5 | Improved |
| add 3D local blocks to all three blocks (ablation) | GaitPart-style backbone · CASIA-B NM/BG/CL (64x44) | rank-1 accuracy: 94.3 -> 97.5 (NM), 90.4 -> 94.3 (BG), 76.7 -> 83.7 (CL) | Improved |
| insert local blocks in higher layers rather than lower layers (ablation) | 3D local CNN · CASIA-B CL (64x44) | rank-1 accuracy: 77.7 -> 81.6 | Improved |
| use Mixture instead of Gaussian sampling (ablation) | 3D local CNN · CASIA-B CL (64x44) | rank-1 accuracy: 83.2 -> 83.7 | No clear effect |
| use Mixture instead of Trilinear sampling (ablation) | 3D local CNN · CASIA-B CL (64x44) | rank-1 accuracy: 82.9 -> 83.7 | No clear effect |
| use Mixture instead of Trilinear sampling (ablation) | 3D local CNN · CASIA-B NM (64x44) | rank-1 accuracy: 96.2 -> 97.5 | Mixed |
| add head local path (ablation) | GaitPart-style backbone · CASIA-B CL (64x44) | rank-1 accuracy: 76.7 -> 77.9 | Improved |
| add arm and torso local paths (ablation) | 3D local CNN · CASIA-B CL (64x44) | rank-1 accuracy: 77.9 -> 79.8 | Improved |
| add left leg path (ablation) | 3D local CNN · CASIA-B CL (64x44) | rank-1 accuracy: 79.8 -> 81.2 | Improved |
| add right leg path (ablation) | 3D local CNN · CASIA-B CL (64x44) | rank-1 accuracy: 81.2 -> 81.9 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GaitPart — GaitPart: Temporal Part-Based Model for Gait Recognition (CVPR 2020) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- GaitPart: Temporal Part-Based Model for Gait Recognition (CVPR 2020 · similarity 0.90)
- Learning Visual Prompt for Gait Recognition (CVPR 2024 · similarity 0.90)
- Deep Gait Recognition: A Survey (TPAMI 2022 · similarity 0.88)
- Gait Recognition via Effective Global-Local Feature Representation and Local Temporal Aggregation (ICCV 2021 · similarity 0.88)
- SkeletonGait: Gait Recognition Using Skeleton Maps (AAAI 2024 · similarity 0.87)
- LidarGait++: Learning Local Features and Size Awareness from LiDAR Point Clouds for 3D Gait Recognition (CVPR 2025 · similarity 0.86)
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.
