Research map / Video / Action recognition
Group-Aware Contrastive Regression for Action Quality Assessment
Xumin Yu, Yongming Rao, Wenliang Zhao, Jiwen Lu, Jie Zhou
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 |
|---|---|---|---|
| replace MLP regressor with group-aware regression tree (GART) (ablation) | I3D + MLP · MTL-AQA (w/ DD) | Spearman correlation: 0.9381 -> 0.9403 | Improved |
| replace MLP regressor with group-aware regression tree (GART) (ablation) | I3D + MLP · MTL-AQA (w/ DD) | R-l2 (x100): 0.394 -> 0.366 | Improved |
| add contrastive regression (CoRe) with exemplars on top of GART (ablation) | I3D + GART · MTL-AQA (w/ DD) | Spearman correlation: 0.9403 -> 0.9512 | Improved |
| add contrastive regression (CoRe) with exemplars on top of GART (ablation) | I3D + GART · MTL-AQA (w/ DD) | R-l2 (x100): 0.366 -> 0.260 | Improved |
| train and test with degree of difficulty (DD) labels for exemplar selection | CoRe + GART · MTL-AQA | Spearman correlation: 0.9341 -> 0.9512 | Improved |
| train and test with degree of difficulty (DD) labels for exemplar selection | CoRe + GART · MTL-AQA | R-l2 (x100): 0.365 -> 0.260 | Improved |
| replace direct regression with CoRe + GART | I3D + MLP* · AQA-7 average | Spearman correlation: 0.7601 -> 0.8401 | Improved |
| replace direct regression with CoRe + GART | I3D + MLP* · AQA-7 average | R-l2 (x100): 3.20 -> 2.12 | Improved |
| replace direct regression with CoRe + GART | I3D + MLP* · JIGSAWS average | Spearman correlation: 0.65 -> 0.85 | Improved |
| replace direct regression with CoRe + GART | I3D + MLP* · JIGSAWS average | R-l2 (x100): 7.373 -> 4.556 | Improved |
| CoRe + GART versus USDL per class | USDL · AQA-7 Big Snow | Spearman correlation: 0.7109 -> 0.6624 | Hurt |
| increase regression tree depth from small (below 4) or beyond 7 (ablation) | CoRe + GART · AQA-7 Diving | Spearman correlation | Hurt |
Closest papers by meaning
- MoAKE: Toward Unified All-in-One Action Quality Assessment via Mixture of Action Knowledge Experts (ECCV 2026 · similarity 0.90)
- Uncertainty-Aware Score Distribution Learning for Action Quality Assessment (CVPR 2020 · similarity 0.90)
- GAIA: Rethinking Action Quality Assessment for AI-Generated Videos (NeurIPS 2024 · similarity 0.88)
- LucidAction: A Hierarchical and Multi-model Dataset for Comprehensive Action Quality Assessment (NeurIPS 2024 · similarity 0.87)
- FineParser: A Fine-grained Spatio-temporal Action Parser for Human-centric Action Quality Assessment (CVPR 2024 · similarity 0.87)
- VE-Bench: Subjective-Aligned Benchmark Suite for Text-Driven Video Editing Quality Assessment (AAAI 2025 · similarity 0.85)
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.
