Research map / Video / Action recognition
ReynoldsFlow: Physics-Inspired Spatiotemporal Flow Representation for Video Understanding
Yu-Hsi Chen, Ching-Kai Lin, PingKong Huang, Chin-Tien Wu
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 |
|---|---|---|---|
| use ReynoldsFlow representation instead of HSV visualization (ablation) | ReynoldsFlow · UCF101 | accuracy: 0.597 -> 0.714 | Improved |
| use ReynoldsFlow representation instead of HSV visualization (ablation) | ReynoldsFlow · UAVDB | AP50: 0.500 -> 0.895 | Improved |
| use ReynoldsFlow representation instead of HSV visualization (ablation) | ReynoldsFlow · GolfDB | PCE: 0.804 -> 0.812 | Improved |
| add curl-free magnitude channel (ablation) | ReynoldsFlow · UAVDB | AP50: 0.803 -> 0.895 | Improved |
| add curl-free magnitude channel (ablation) | ReynoldsFlow · Anti-UAV | AP50-95: 0.407 -> 0.446 | Improved |
| add curl-free magnitude channel (ablation) | ReynoldsFlow · HMDB51 | accuracy: 0.375 -> 0.402 | Improved |
| put DF magnitude in red and CF in green rather than swapped (ablation) | ReynoldsFlow · UAVDB | AP50: 0.831 -> 0.895 | Improved |
| use ReynoldsFlow input instead of RGB (ablation) | YOLOv11n on UAVDB · UAVDB | AP50: 0.811 -> 0.895 | Improved |
| use ReynoldsFlow input instead of RGB (ablation) | C3D on HMDB51 · HMDB51 | accuracy: 0.372 -> 0.402 | Improved |
| use ReynoldsFlow input instead of RGB (ablation) | YOLOv11n on ARD100 · ARD100 | AP50: 0.554 -> 0.602 | Improved |
| use ReynoldsFlow instead of grayscale/infrared input (ablation) | YOLOv11n on Anti-UAV · Anti-UAV | AP50: 0.781 -> 0.792 | Improved |
| use ReynoldsFlow instead of best DL flow (SEA-RAFT) for action recognition (ablation) | C3D on HMDB51 · HMDB51 | accuracy: 0.419 -> 0.402 | Hurt |
Closest papers by meaning
- FlowFeat: Pixel-Dense Embedding of Motion Profiles (NeurIPS 2025 · similarity 0.87)
- DeltaFlow: An Efficient Multi-frame Scene Flow Estimation Method (NeurIPS 2025 · similarity 0.87)
- Semantic Flow: Learning Semantic Fields of Dynamic Scenes from Monocular Videos (ICLR 2024 · similarity 0.87)
- StreamFlow: Streamlined Multi-Frame Optical Flow Estimation for Video Sequences (NeurIPS 2024 · similarity 0.87)
- E-MoFlow: Learning Egomotion and Optical Flow from Event Data via Implicit Regularization (NeurIPS 2025 · similarity 0.86)
- Taming generative video models for zero-shot optical flow extraction (NeurIPS 2025 · 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.
