Unsupervised learning of monocular depth and ego-motion with space-temporal-centroid loss.
Junning ZhangQunxing SuPengyuan LiuChao XuYanlong ChenPublished in: Int. J. Mach. Learn. Cybern. (2020)
Keyphrases
- ego motion
- unsupervised learning
- field of view
- stereo camera
- space time
- image sequences
- d scene
- optical flow
- camera motion
- dynamic scenes
- image velocity
- visual odometry
- depth information
- flow field
- motion parameters
- robust estimation
- multi camera
- spatio temporal
- depth map
- motion detection
- video sequences
- single camera
- moving camera
- optic flow
- stereo vision
- vision system
- reinforcement learning
- velocity field
- face recognition
- computer vision
- machine learning
- high quality
- principal component analysis
- feature space
- semi supervised
- least squares
- medical images