Research map / Video / Action recognition
Novel Motion Patterns Matter for Practical Skeleton-Based Action Recognition
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 |
|---|---|---|---|
| Use Mask-GCN with local policy-network modeling instead of CTR-GCN | skeleton-based action recognition · Novel-motion dataset, CS 1 | accuracy: 46.90 -> 51.05 | Helped |
| Use Mask-GCN with local policy-network modeling instead of CTR-GCN | skeleton-based action recognition · Novel-motion dataset, CS 2 | accuracy: 62.48 -> 66.90 | Helped |
| Use Mask-GCN with local policy-network modeling instead of CTR-GCN | skeleton-based action recognition · Novel-motion dataset, CV 1 | accuracy: 44.46 -> 57.79 | Helped |
| Use Mask-GCN with local policy-network modeling instead of CTR-GCN | skeleton-based action recognition · Novel-motion dataset, CV 2 | accuracy: 66.23 -> 73.00 | Helped |
| Use local rather than global policy-network modeling (ablation) | Mask-GCN · Novel-motion dataset, CS 2 | accuracy: 65.79 -> 66.90 | Helped |
| Use local rather than global policy-network modeling (ablation) | Mask-GCN · Novel-motion dataset, CV 1 | accuracy: 45.53 -> 57.79 | Helped |
| Use local rather than global policy-network modeling (ablation) | Mask-GCN · Novel-motion dataset, CV 2 | accuracy: 68.81 -> 73.00 | Helped |
| Combine local and global modeling rather than use local modeling alone (ablation) | Mask-GCN · Novel-motion dataset, CS 2 | accuracy: 66.90 -> 64.84 | Hurt |
| Combine local and global modeling rather than use local modeling alone (ablation) | Mask-GCN · Novel-motion dataset, CV 2 | accuracy: 73.00 -> 72.08 | Hurt |
| Set Gumbel Softmax temperature to 0.1 instead of 0.01 (ablation) | Mask-GCN policy network · Novel-motion dataset | accuracy: 50.32 -> 51.05 | Hurt |
| Set Gumbel Softmax temperature to 0.001 instead of 0.01 (ablation) | Mask-GCN policy network · Novel-motion dataset | accuracy: 51.05 -> 33.52 | Hurt |
| Add an adjacency-matrix mask generation block (ablation) | Mask-GCN · Novel-motion dataset, CS 1 and CS 2 | accuracy | Helped |
Closest papers by meaning
- Actional-Structural Graph Convolutional Networks for Skeleton-Based Action Recognition (CVPR 2019 · similarity 0.93)
- Learning Graph Convolutional Network for Skeleton-Based Human Action Recognition by Neural Searching (AAAI 2020 · similarity 0.92)
- Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition (CVPR 2025 · similarity 0.92)
- Skeleton-Based Action Recognition With Shift Graph Convolutional Network (CVPR 2020 · similarity 0.91)
- BlockGCN: Redefine Topology Awareness for Skeleton-Based Action Recognition (CVPR 2024 · similarity 0.90)
- Dynamic Semantic-Based Spatial Graph Convolution Network for Skeleton-Based Human Action Recognition (AAAI 2024 · similarity 0.90)
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.