Multi-view human action recognition using 2D motion templates based on MHIs and their HOG description.
Fiza MurtazaMuhammad Haroon YousafSergio A. VelastinPublished in: IET Comput. Vis. (2016)
Keyphrases
- multi view
- three dimensional
- single view
- multiple views
- depth map
- dynamic scenes
- d objects
- object detection
- semi supervised
- action recognition
- image sequences
- camera calibration
- view synthesis
- scene reconstruction
- multi view learning
- range images
- multi view reconstruction
- multiple viewpoints
- multi view stereo
- motion estimation
- surface reconstruction
- multi view images
- co training
- pedestrian detection
- image motion
- bundle adjustment
- machine learning
- d scene
- free viewpoint
- feature set
- optical flow
- multi view face detection