Research map / Video / Video segmentation and tracking
Decoupling and Recoupling Spatiotemporal Representation for RGB-D-Based Motion Recognition
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 |
|---|---|---|---|
| use the full RGB-D method with CAPF rather than the strongest listed baseline | Decoupled and Recoupled Spatiotemporal Representation · NvGesture, RGB+Depth | accuracy: 88.38 -> 91.70 | Helped |
| use the full RGB-D method with CAPF rather than the strongest listed baseline | Decoupled and Recoupled Spatiotemporal Representation · THU-READ, RGB+Depth | accuracy: 84.90 -> 87.04 | Helped |
| use the full RGB-D method with CAPF rather than the strongest listed baseline | Decoupled and Recoupled Spatiotemporal Representation · Chalearn IsoGD, RGB+Depth | accuracy: 65.54 -> 66.79 | Helped |
| use the full RGB-D method rather than CTR-GCN | Decoupled and Recoupled Spatiotemporal Representation · NTU-RGBD CS | accuracy: 92.4 -> 94.2 | Helped |
| add FRP to the RGB model (ablation) | RGB unimodal network · NvGesture | accuracy: 85.21 -> 86.67 | Helped |
| add FRP to the RGB model (ablation) | RGB unimodal network · THU-READ | accuracy: 75.24 -> 78.33 | Helped |
| add multi-loss constraints to DTN branches (ablation) | RGB unimodal network · NvGesture | accuracy: 86.67 -> 87.08 | Helped |
| add multi-loss constraints to DTN branches (ablation) | RGB unimodal network · THU-READ | accuracy: 78.33 -> 80.53 | Helped |
| add the sharpness constraint (ablation) | RGB unimodal network · NvGesture | accuracy: 87.08 -> 89.13 | Helped |
| add the sharpness constraint (ablation) | RGB unimodal network · THU-READ | accuracy: 80.53 -> 80.91 | Helped |
| add k-NN attention to temporal representation (ablation) | RGB unimodal network · NvGesture | accuracy: 89.13 -> 89.58 | Helped |
| add k-NN attention to temporal representation (ablation) | RGB unimodal network · THU-READ | accuracy: 80.91 -> 81.67 | Helped |
Closest papers by meaning
- Spatiotemporal Fusion in 3D CNNs: A Probabilistic View (CVPR 2020 · similarity 0.88)
- Wavelet-Driven Spatiotemporal Predictive Learning: Bridging Frequency and Time Variations (AAAI 2024 · similarity 0.85)
- Building a Multi-modal Spatiotemporal Expert for Zero-shot Action Recognition with CLIP (AAAI 2025 · similarity 0.84)
- UniSTD: Towards Unified Spatio-Temporal Learning across Diverse Disciplines (CVPR 2025 · similarity 0.84)
- Bridge Frame and Event: Common Spatiotemporal Fusion for High-Dynamic Scene Optical Flow (CVPR 2025 · similarity 0.84)
- RGB-Only Supervised Camera Parameter Optimization in Dynamic Scenes (NeurIPS 2025 · similarity 0.84)
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.