Research map / Video / Action recognition
Intra- and Inter-Action Understanding via Temporal Action Parsing
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 TransParser instead of the strongest non-human parsing baseline | Temporal action parsing · TAPOS | avg. F1-score (rel.): 0.3502 -> 0.4745 | Helped |
| Use TransParser instead of the strongest non-human parsing baseline | Temporal action parsing · TAPOS | avg. F1-score (abs.): 0.3483 -> 0.3981 | Helped |
| Sample TSN frames using TransParser parsing rather than uniform sampling | TSN · TAPOS action classification, RGB | Top-1 Acc.: 83.97 -> 84.80 | Helped |
| Sample TSN frames using TransParser parsing rather than uniform sampling | TSN · TAPOS action classification, RGB + Flow | Top-1 Acc.: 91.01 -> 91.62 | Helped |
| Align TSN segments to annotated sub-actions instead of sampling uniformly | TSN · TAPOS action classification, RGB | Top-1 Acc.: 83.97 -> 88.83 | Helped |
| Add the local semantic loss (ablation) | TransParser with one SPS unit · TAPOS | avg. F1: 0.2897 -> 0.3996 | Helped |
| Add a second SPS unit (ablation) | TransParser with local loss · TAPOS | avg. F1: 0.3996 -> 0.4210 | Helped |
| Increase SPS units beyond two (ablation) | TransParser · TAPOS | avg. F1 | No clear effect |
Closest papers by meaning
- ActionAtlas: A VideoQA Benchmark for Domain-specialized Action Recognition (NeurIPS 2024 · similarity 0.88)
- Unsupervised Action Segmentation via Fast Learning of Semantically Consistent Actoms (AAAI 2024 · similarity 0.88)
- Action Detail Matters: Refining Video Recognition with Local Action Queries (CVPR 2025 · similarity 0.87)
- FineSports: A Multi-person Hierarchical Sports Video Dataset for Fine-grained Action Understanding (CVPR 2024 · similarity 0.87)
- ViSTec: Video Modeling for Sports Technique Recognition and Tactical Analysis (AAAI 2024 · similarity 0.87)
- FineParser: A Fine-grained Spatio-temporal Action Parser for Human-centric Action Quality Assessment (CVPR 2024 · 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.