Research map / Vision-language and multimodal / Cross-modal learning
Exploiting Frequency Dynamics for Enhanced Multimodal Event-based Action Recognition
Meiqi Cao, Xiangbo Shu, Xin Jiang, Rui Yan, Yazhou Yao, Jinhui Tang
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 Cross-modal Frequency Enhancer (ablation) | EMP baseline · SeAct | Top-1 accuracy: 71.55 -> 72.41 | Improved |
| add Cross-modal Frequency Enhancer (ablation) | EMP baseline · DVS128 Gesture | Top-1 accuracy: 97.34 -> 97.72 | Improved |
| add High-frequency Guided Selector (ablation) | EMP baseline · SeAct | Top-1 accuracy: 71.55 -> 73.27 | Improved |
| add High-frequency Guided Selector (ablation) | EMP baseline · DVS128 Gesture | Top-1 accuracy: 97.34 -> 98.10 | Improved |
| combine CFE and HGS (ablation) | EMP baseline · SeAct | Top-1 accuracy: 71.55 -> 75.00 | Improved |
| combine CFE and HGS (ablation) | EMP baseline · DVS128 Gesture | Top-1 accuracy: 97.34 -> 98.86 | Improved |
| select 60% of tokens instead of keeping all (mu 1.0 -> 0.6) (ablation) | HGS · SeAct | Top-1 accuracy: 72.41 -> 75.00 | Improved |
| insert token selection at layer 8 instead of layer 4 (ablation) | HGS · SeAct | Top-1 accuracy: 73.27 -> 75.00 | Improved |
| add RGB frames to event streams | C3D · HARDVS | Top-1 accuracy: 49.94 -> 50.88 | Improved |
| add RGB frames to event streams | TimeSformer · HARDVS | Top-1 accuracy: 50.77 -> 51.57 | Improved |
| use EMP instead of EventCrab | EventCrab · PAF | Top-1 accuracy: 96.49 -> 99.80 | Improved |
| use EMP instead of SpikMamba | SpikMamba · DVS128 Gesture | Top-1 accuracy: 99.01 -> 98.86 | Hurt |
Closest papers by meaning
- EventDance: Unsupervised Source-free Cross-modal Adaptation for Event-based Object Recognition (CVPR 2024 · similarity 0.87)
- FlexEvent: Towards Flexible Event-Frame Object Detection at Varying Operational Frequencies (NeurIPS 2025 · similarity 0.87)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 2025 · similarity 0.86)
- Adapting to the Unknown: Training-Free Audio-Visual Event Perception with Dynamic Thresholds (CVPR 2025 · similarity 0.86)
- Cross-modal Multi-task Learning for Multimedia Event Extraction (AAAI 2025 · similarity 0.85)
- ExACT: Language-guided Conceptual Reasoning and Uncertainty Estimation for Event-based Action Recognition and More (CVPR 2024 · similarity 0.85)
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.
