Biologically-inspired robust motion segmentation using mutual information.
Anna-Louise EllisJames M. FerrymanPublished in: Comput. Vis. Image Underst. (2014)
Keyphrases
- biologically inspired
- motion segmentation
- mutual information
- multibody
- moving objects
- receptive fields
- motion estimation
- image sequences
- missing data
- motion model
- epipolar geometry estimation
- humanoid robot
- image registration
- biologically plausible
- model selection
- dynamic scenes
- feature tracking
- scene classification
- motion field
- optical flow
- similarity measure
- feature selection
- spiking neural networks
- computer vision
- machine learning
- structure from motion
- geometric properties
- video sequences
- multiple moving objects
- feature trajectories