SP-Flow: Self-supervised optical flow correspondence point prediction for real-time SLAM.
Zixuan QinMengxiao YinGuiqing LiFeng YangPublished in: Comput. Aided Geom. Des. (2020)
Keyphrases
- optical flow
- real time
- visual odometry
- flow field
- visual slam
- image sequences
- mobile robot
- prediction accuracy
- flow estimation
- simultaneous localization and mapping
- camera tracking
- camera motion
- prediction model
- dense correspondence
- least squares
- optical flow estimation
- motion discontinuities
- control system
- camera pose
- monocular slam
- monocular camera
- flow vectors
- autonomous navigation
- motion field
- motion estimation
- vision system
- dynamic environments
- d scene
- prediction error
- optical flow computation
- motion parameters
- ego motion
- vector field
- bundle adjustment
- image matching
- indoor environments
- real time tracking
- point to point correspondences
- point tracking
- velocity field