InterpoNet, a Brain Inspired Neural Network for Optical Flow Dense Interpolation.
Shay ZweigLior WolfPublished in: CVPR (2017)
Keyphrases
- optical flow
- neural network
- dense optical flow
- optical flow fields
- scene flow
- motion field estimation
- optical flow computation
- optical flow field
- motion field
- image sequences
- flow estimation
- optical flow estimation
- back propagation
- neural network model
- flow field
- d scene
- artificial neural networks
- computer vision
- dense correspondence
- least squares
- point tracking
- fuzzy logic
- recurrent neural networks
- motion estimation
- velocity field
- motion model
- motion segmentation
- human brain
- network architecture
- neural network is trained
- image interpolation
- interpolation method
- self organizing maps
- moving objects
- dense optical flow estimation
- dense motion estimation
- brain activity
- functional magnetic resonance imaging
- displacement field
- ego motion
- motion analysis
- vector field
- super resolution
- three dimensional
- genetic algorithm