Research map / Video / Action recognition
Beyond Label Semantics: Language-Guided Action Anatomy for Few-shot Action Recognition
Zefeng Qian, Xincheng Yao, Yifei Huang, Chongyang Zhang, Jiangyong Ying, Hong Sun
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 |
|---|---|---|---|
| add visual anatomy (atomic action segmentation) (ablation) | CLIP-FSAR · HMDB51 1-shot | accuracy: 75.8 -> 79.9 | Improved |
| add textual anatomy (atomic descriptions) (ablation) | CLIP-FSAR · HMDB51 1-shot | accuracy: 75.8 -> 79.6 | Improved |
| add visual anatomy alone (ablation) | CLIP-FSAR · HMDB51 5-shot | accuracy: 87.7 -> 86.0 | Hurt |
| add both visual and textual anatomy (ablation) | LGA · HMDB51 1-shot | accuracy: 75.8 -> 80.8 | Improved |
| add video-text matching (ablation) | LGA · HMDB51 1-shot | accuracy: 80.8 -> 86.8 | Improved |
| add video-text matching (ablation) | LGA · HMDB51 5-shot | accuracy: 88.2 -> 89.3 | Improved |
| use atomic descriptions instead of action label with self-attention global fusion (ablation) | fusion module · HMDB51 1-shot | accuracy: 79.3 -> 85.2 | Improved |
| use cross-attention instead of self-attention fusion (ablation) | fusion module · HMDB51 1-shot | accuracy: 79.3 -> 81.7 | Improved |
| use local instead of global fusion with atomic descriptions (ablation) | fusion module · SSv2-Small 5-shot | accuracy: 65.4 -> 66.5 | Mixed |
| use AB-MHM instead of Bi-MHM for video-video matching (ablation) | LGA · SSv2-Small 1-shot | accuracy: 55.2 -> 58.9 | Improved |
| use CLUSTER-Segment instead of TW-FINCH (ablation) | Visual Anatomy · HMDB51 1-shot | accuracy: 82.9 -> 86.8 | Improved |
| use CLUSTER-Segment instead of uniform HARD split (ablation) | Visual Anatomy · SSv2-Small 1-shot | accuracy: 57.0 -> 58.9 | Improved |
Closest papers by meaning
- Storyboard-guided Alignment for Fine-grained Video Action Recognition (NeurIPS 2025 · similarity 0.89)
- SeFAR: Semi-supervised Fine-grained Action Recognition with Temporal Perturbation and Learning Stabilization (AAAI 2025 · similarity 0.89)
- LLaVAction: evaluating and training multi-modal large language models for action understanding (ICLR 2026 · similarity 0.88)
- Temporal Alignment-Free Video Matching for Few-shot Action Recognition (CVPR 2025 · similarity 0.88)
- Part-aware Unified Representation of Language and Skeleton for Zero-shot Action Recognition (CVPR 2024 · similarity 0.88)
- Frame Order Matters: A Temporal Sequence-Aware Model for Few-Shot Action Recognition (AAAI 2025 · similarity 0.88)
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.
