Maximum-likelihood object tracking from multi-view video by combining homography and epipolar constraints.
Yixiao YunIrene Yu-Hua GuHamid K. AghajanPublished in: ICDSC (2012)
Keyphrases
- multi view
- object tracking
- video sequences
- epipolar constraint
- video surveillance
- camera motion
- multiple views
- dynamic scenes
- fundamental matrix
- epipolar geometry
- multiple cameras
- image sequences
- feature points
- single view
- particle filter
- depth map
- appearance model
- video data
- moving objects
- mean shift
- video frames
- d objects
- point correspondences
- moving camera
- three dimensional
- motion estimation
- camera calibration
- photometric stereo
- semi supervised
- structure from motion
- multi camera
- motion vectors
- computer vision
- motion field
- machine learning
- stereo camera
- d scene
- surface reconstruction
- multiscale
- motion parameters
- range images
- image pairs
- single image
- facial expressions
- stereo correspondence