Research map / Video / Action recognition
Task-Specific Distance Correlation Matching for Few-Shot Action Recognition
Fei Long, Yao Zhang, Jiaming Lv, Jiangtao Xie, Peihua Li
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 Ladder Side Network to zero-shot CLIP (ablation) | CLIP ViT-B/16 · SSv2-Full 1-shot | accuracy (%): 37.0 -> 67.1 | Improved |
| add IF-DaC on top of LSN (ablation) | TS-FSAR · SSv2-Full 1-shot | accuracy (%): 67.1 -> 71.4 | Improved |
| add TSM on top of LSN+IF-DaC (ablation) | TS-FSAR · SSv2-Full 1-shot | accuracy (%): 71.4 -> 73.8 | Improved |
| add GLAC on top of LSN+IF-DaC+TSM (ablation) | TS-FSAR · HMDB51 5-shot | accuracy (%): 85.6 -> 88.9 | Improved |
| use query-specific task prototype (average) (ablation) | TS-FSAR · SSv2-Full 1-shot | accuracy (%): 74.1 -> 75.1 | Improved |
| use averaging instead of cross-attention for query-specific prototype (ablation) | TS-FSAR · SSv2-Full 1-shot | accuracy (%): 74.1 -> 75.1 | Improved |
| use alpha-DC instead of cosine similarity for inter-frame dependency (ablation) | TS-FSAR · SSv2-Full 1-shot | accuracy (%): 71.5 -> 75.1 | Improved |
| use TS-DCM instead of BiMHM (ablation) | TS-FSAR · SSv2-Full 1-shot | accuracy (%): 72.3 -> 75.1 | Improved |
| add IF-DaC to GAP metric (ablation) | GAP · SSv2-Full 1-shot | accuracy (%): 68.6 -> 72.0 | Improved |
| use 12-layer LSN instead of 3-layer (ablation) | TS-FSAR · SSv2-Full 1-shot | accuracy (%): 67.0 -> 75.1 | Improved |
| TS-FSAR versus TSAM on Kinetics-100 | TS-FSAR · Kinetics-100 5-shot | accuracy (%): 97.1 -> 96.6 | Hurt |
| TS-FSAR versus TSAM on SSv2-Full | TS-FSAR · SSv2-Full 1-shot | accuracy (%): 65.8 -> 75.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- TSAM — TSAM: Temporal SAM Augmented with Multimodal Prompts for Referring Audio-Visual Segmentation (CVPR 2025) · better on 6 of 10 shared results
- D2ST-Adapter — D2ST-Adapter: Disentangled-and-Deformable Spatio-Temporal Adapter for Few-shot Action Recognition (ICCV 2025) · better on 10 of 10 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Hybrid Relation Guided Set Matching for Few-Shot Action Recognition (CVPR 2022 · similarity 0.89)
- Temporal Alignment-Free Video Matching for Few-shot Action Recognition (CVPR 2025 · similarity 0.89)
- Frame Order Matters: A Temporal Sequence-Aware Model for Few-Shot Action Recognition (AAAI 2025 · similarity 0.88)
- BDC-CLIP: Brownian Distance Covariance for Adapting CLIP to Action Recognition (ICML 2025 · similarity 0.87)
- TAMT: Temporal-Aware Model Tuning for Cross-Domain Few-Shot Action Recognition (CVPR 2025 · similarity 0.87)
- Pushing the Limit of Fine-Tuning for Few-Shot Learning: Where Feature Reusing Meets Cross-Scale Attention (AAAI 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.
