Pairwise Decomposition of Image Sequences for Active Multi-View Recognition.
Edward JohnsStefan LeuteneggerAndrew J. DavisonPublished in: CoRR (2016)
Keyphrases
- multi view
- pairwise
- image sequences
- three dimensional
- semi supervised
- depth map
- single view
- multiple views
- cross view action recognition
- face alignment
- d objects
- object recognition
- multiple cameras
- camera calibration
- feature extraction
- multiple viewpoints
- dynamic scenes
- video sequences
- scene reconstruction
- object tracking
- multi view clustering
- viewpoint
- markov random field
- structure from motion
- multi view face detection
- surface reconstruction
- multi view images
- bundle adjustment
- multi view reconstruction
- multi view learning
- co training
- motion estimation
- light field
- visual hull
- moving objects
- action recognition
- view synthesis
- similarity measure
- unlabeled data
- high resolution
- human motion
- depth video
- belief propagation