An Effective Temporal Localization Method with Multi-View 3D Action Recognition for Untrimmed Naturalistic Driving Videos.
Manh Tung TranMinh Quan VuNgoc Duong HoangKhac-Hoai Nam BuiPublished in: CVPR Workshops (2022)
Keyphrases
- multi view
- view invariant
- action recognition
- single view
- human actions
- camera calibration
- dynamic scenes
- three dimensional
- multiple views
- human activities
- depth map
- semi supervised
- spatio temporal
- bag of words
- activity recognition
- d objects
- human motion
- view synthesis
- computer vision
- space time
- multiple cameras
- viewpoint
- range images
- multiple viewpoints
- object recognition
- video sequences
- feature extraction