Research map / Video / Action recognition
Quo Vadis, Skeleton Action Recognition?
Pranay Gupta, Anirudh Thatipelli, Aditya Aggarwal, Shubh Maheshwari, Neel Trivedi, Sourav Das, Ravi Kiran Sarvadevabhatla
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 |
|---|---|---|---|
| average-pool ensemble of best-5 models | MS-G3D / 4s-ShiftGCN · NTU-120 Cross Setup | accuracy: 87.65 -> 88.80 | Improved |
| average-pool ensemble of best-5 models | MS-G3D · NTU-120 Cross Subject | accuracy: 85.92 -> 87.22 | Improved |
| pretrain on NTU-120 VIBE skeletons then finetune | 4s-ShiftGCN · Skeletics-152 | accuracy: 56.15 -> 57.01 | Improved |
| pretrain on NTU-120 VIBE skeletons then finetune | MS-G3D · Skeletics-152 | accuracy: 56.39 -> 55.75 | Hurt |
| train on full Skeletics-152 rather than only the 23 mimetics classes (ablation) | MS-G3D · Skeleton-Mimetics | accuracy: 49.22 -> 57.37 | Improved |
| train on full Skeletics-152 rather than only the 23 mimetics classes (ablation) | 4s-ShiftGCN · Skeleton-Mimetics | accuracy: 51.10 -> 56.11 | Improved |
| train on in-the-wild Skeletics-152 instead of lab NTU-120 | 4s-ShiftGCN · Metaphorics | mean cosine similarity: 0.04 -> 0.08 | Improved |
| train on in-the-wild Skeletics-152 instead of lab NTU-120 | MS-G3D · Metaphorics | mean cosine similarity: 0.14 -> 0.10 | Hurt |
| move from lab NTU-120 to in-the-wild Skeletics-152 | 4s-ShiftGCN · accuracy | accuracy: 87.65 -> 57.01 | Hurt |
Closest papers by meaning
- Action Recognition Using Rate-Invariant Analysis of Skeletal Shape Trajectories (TPAMI 2016 · similarity 0.89)
- InfoGCN: Representation Learning for Human Skeleton-Based Action Recognition (CVPR 2022 · similarity 0.88)
- Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition (CVPR 2025 · similarity 0.88)
- Recovering Complete Actions for Cross-dataset Skeleton Action Recognition (NeurIPS 2024 · similarity 0.88)
- Skeleton-based Action Recognition with Non-linear Dependency Modeling and Hilbert-Schmidt Independence Criterion (AAAI 2025 · similarity 0.87)
- Heterogeneous Skeleton-Based Action Representation Learning (CVPR 2025 · 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.
