Back to Basics: Unsupervised Learning of Optical Flow via Brightness Constancy and Motion Smoothness.
Jason J. YuAdam W. HarleyKonstantinos G. DerpanisPublished in: CoRR (2016)
Keyphrases
- optical flow
- brightness constancy
- unsupervised learning
- motion boundaries
- brightness constancy assumption
- optical flow computation
- motion model
- optical flow estimation
- smoothness constraint
- motion field
- spatial smoothness
- motion estimation
- image sequences
- flow estimation
- optical flow algorithm
- motion segmentation
- moving objects
- motion parameters
- velocity field
- motion analysis
- camera motion
- feature tracking
- flow field
- ego motion
- d scene
- dimensionality reduction
- image motion
- semi supervised
- computer vision
- consecutive frames
- vector field
- particle filter
- image derivatives
- multi frame
- space time
- spatio temporal
- image intensity