Research map / Label-efficient and robust learning / Recognition and long-tail
Active Inference for Micro-Gesture Recognition: EFE-Guided Temporal Sampling and Adaptive Learning
Weijia Feng, Jingyu Yang, Ruojia Zhang, Fengtao Sun, Qian Gao, Chenyang Wang, Tongtong Su, Jia Guo, Xiaobai Li, Minglai Shao
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 uncertainty-aware augmentation (UMIX) (ablation) | ResNet-50 baseline · SMG | accuracy (%): 50.49 -> 57.54 | Improved |
| add EFE-guided temporal selection (ablation) | ResNet-50 baseline · SMG | accuracy (%): 50.49 -> 56.40 | Improved |
| add EFE-guided spatial selection (ablation) | ResNet-50 baseline · SMG | accuracy (%): 50.49 -> 55.40 | Improved |
| combine UMIX, temporal selection and spatial selection (ablation) | ResNet-50 baseline · SMG | accuracy (%): 50.49 -> 63.47 | Improved |
| replace Logic-in-Frames keyframe selection with EFE-guided selection (ablation) | UAAI · SMG | accuracy (%): 61.31 -> 63.47 | Improved |
| replace Video Tree keyframe selection with EFE-guided selection (ablation) | UAAI · SMG | accuracy (%): 59.51 -> 63.47 | Improved |
| increase Monte Carlo dropout samples from 2 to 5 (ablation) | UMIX uncertainty estimation · SMG | validation accuracy and convergence | Improved |
| increase Monte Carlo dropout samples from 5 to 8 (ablation) | UMIX uncertainty estimation · SMG | validation accuracy | No clear effect |
| increase Monte Carlo dropout samples from 2 to 8 (ablation) | UMIX uncertainty estimation · SMG | training wall-clock time (s): ~152 -> ~255 | Hurt |
| use RGB-only UAAI versus skeleton-based MS-G3D | UAAI · SMG | accuracy (%): 64.75 -> 63.47 | Hurt |
Closest papers by meaning
- Revisiting Noise Resilience Strategies in Gesture Recognition: Short-Term Enhancement in sEMG Analysis (ICML 2025 · similarity 0.86)
- SpGesture: Source-Free Domain-adaptive sEMG-based Gesture Recognition with Jaccard Attentive Spiking Neural Network (NeurIPS 2024 · similarity 0.85)
- PULSE: A Synchronized Five-Modality Dataset for Sensorimotor Coordination in Long-Horizon Daily Activities (NeurIPS 2026 · similarity 0.84)
- emg2pose: A Large and Diverse Benchmark for Surface Electromyographic Hand Pose Estimation (NeurIPS 2024 · similarity 0.84)
- emg2qwerty: A Large Dataset with Baselines for Touch Typing using Surface Electromyography (NeurIPS 2024 · similarity 0.83)
- Multi-Modal Hand-to-Mouth Gesture Recognition in Activity-Oriented RGB-Thermal Footage (Student Abstract) (AAAI 2025 · similarity 0.82)
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.
