Deep-learning Optical Flow Outperforms PIV in Obtaining Velocity Fields from Active Nematics.
Phu N. TranSattvic RayLinnea LemmaYunrui LiReef SweeneyAparna BaskaranZvonimir DogicPengyu HongMichael F. HaganPublished in: CoRR (2024)
Keyphrases
- velocity field
- deep learning
- optical flow
- flow field
- vector field
- optical flow estimation
- image sequences
- unsupervised learning
- image velocity
- motion field
- machine learning
- ego motion
- motion estimation
- flow estimation
- motion model
- mental models
- moving objects
- weakly supervised
- computer vision
- camera motion
- motion segmentation
- d objects
- viewpoint
- motion parameters
- object recognition