Unsupervised Estimation of Monocular Depth and VO in Dynamic Environments via Hybrid Masks.
Qiyu SunYang TangChongzhen ZhangChaoqiang ZhaoFeng QianJürgen KurthsPublished in: IEEE Trans. Neural Networks Learn. Syst. (2022)
Keyphrases
- dynamic environments
- visual slam
- depth cues
- mobile robot
- depth from motion
- visual sensor
- autonomous navigation
- path planning
- autonomous agents
- image sequences
- agent systems
- visual odometry
- depth perception
- depth map
- optical flow
- changing environment
- edge detection
- real environment
- potential field
- pose estimation
- unsupervised learning
- collision avoidance
- depth information
- stereo camera
- highly dynamic environments
- real time
- semi supervised
- reinforcement learning agents
- agent based systems
- belief space
- stereo matching