Specificity and Latent Correlation Learning for Action Recognition Using Synthetic Multi-View Data From Depth Maps.
Bin LiangLihong ZhengPublished in: IEEE Trans. Image Process. (2017)
Keyphrases
- multi view
- depth map
- multi view learning
- single view
- multiple views
- three dimensional
- high quality
- data sets
- training data
- stereo matching
- view synthesis
- d objects
- depth information
- multi view stereo
- depth data
- input data
- multiple cameras
- learning algorithm
- depth estimation
- real scenes
- range images
- multi view images
- free viewpoint
- image data
- data points
- learning process
- viewpoint
- depth video
- camera parameters
- dynamic scenes
- photometric stereo
- d scene
- camera calibration
- single image
- input image
- video sequences