Self-supervised monocular depth estimation in dynamic scenes with moving instance loss.
Min YueGuangyuan FuMing WuXin ZhangHongyang GuPublished in: Eng. Appl. Artif. Intell. (2022)
Keyphrases
- dynamic scenes
- depth estimation
- moving objects
- moving camera
- ego motion
- image sequences
- space time
- multi view
- background subtraction
- motion segmentation
- video sequences
- multiple cameras
- pose estimation
- optical flow
- spatio temporal
- motion estimation
- real time
- multiple views
- frame rate
- motion model
- viewpoint
- machine learning
- computer vision
- video surveillance
- stereo vision