Enhancing self-supervised monocular depth estimationwith traditional visual odometry.
Lorenzo AndraghettiPanteleimon MyriokefalitakisPier Luigi DovesiBelén LuqueMatteo PoggiAlessandro PieropanStefano MattocciaPublished in: CoRR (2019)
Keyphrases
- visual odometry
- depth images
- ego motion
- depth information
- long range
- autonomous navigation
- field of view
- position information
- kalman filtering
- depth map
- optical flow
- pose estimation
- camera motion
- stereo camera
- robust estimation
- real time
- mobile robot
- simultaneous localization and mapping
- stereo vision
- optic flow
- camera pose
- d scene
- range data
- low resolution
- least squares
- kalman filter
- d objects
- path planning
- high resolution
- computer vision