Research map / Video / Action recognition
Few-Shot Video Classification via Representation Fusion and Promotion Learning
Haifeng Xia, Kai Li, Martin Renqiang Min, Zhengming Ding
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 RFPL (MAL + RIR) to OTAM | OTAM · SSv2-Full 5-way 5-shot | accuracy: 52.3 -> 61.0 | Improved |
| add RFPL to OTAM | OTAM · UCF101 5-way 1-shot | accuracy: 79.9 -> 84.3 | Improved |
| add RFPL to TRX | TRX · SSv2-Full 5-way 5-shot | accuracy: 64.6 -> 64.6 | No clear effect |
| add RFPL to TRX | TRX · UCF101 5-way 5-shot | accuracy: 96.1 -> 96.3 | Improved |
| add RFPL to STRM | STRM · UCF101 5-way 5-shot | accuracy: 96.9 -> 96.7 | Hurt |
| add RFPL to STRM | STRM · Kinetics 5-way 5-shot | accuracy: 86.7 -> 87.7 | Improved |
| add RFPL to STRM | STRM · SSv2-Full 5-way 5-shot | accuracy: 68.1 -> 69.5 | Improved |
| increase fusion weight rho to 0.5 (ablation) | RFPL meta-action learning · SSv2-Full val | accuracy | Hurt |
| add meta-action learning alone (Ours-v1) or reinforced image representation alone (Ours-v2) to OTAM (ablation) | OTAM · Kinetics 2/3/4-shot | accuracy | Improved |
| combine MAL and RIR rather than either alone (ablation) | OTAM · Kinetics 2/3/4-shot | accuracy | Improved |
Closest papers by meaning
- Unsupervised Few-Shot Action Recognition via Action-Appearance Aligned Meta-Adaptation (ICCV 2021 · similarity 0.89)
- Domain-Specific Priors and Meta Learning for Few-Shot First-Person Action Recognition (TPAMI 2021 · similarity 0.88)
- A Comprehensive Review of Few-Shot Action Recognition (IJCV 2025 · similarity 0.88)
- Frame-Wise Action Representations for Long Videos via Sequence Contrastive Learning (CVPR 2022 · similarity 0.88)
- Few-Shot Action Recognition with Hierarchical Matching and Contrastive Learning (ECCV 2022 · similarity 0.88)
- Pushing the Limit of Fine-Tuning for Few-Shot Learning: Where Feature Reusing Meets Cross-Scale Attention (AAAI 2024 · 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.
