Stereo dense depth tracking based on optical flow using frames and events.
Antea HadvigerIvan MarkovicIvan PetrovicPublished in: Adv. Robotics (2021)
Keyphrases
- optical flow
- consecutive frames
- single frame
- stereo disparity
- motion model
- dense optical flow
- feature tracking
- motion segmentation
- optical flow estimation
- computer vision
- multi frame
- occlusion detection
- moving objects
- image sequences
- motion cues
- rigid body motion
- lucas kanade
- optical flow computation
- motion trajectories
- point tracking
- optical flow algorithm
- motion features
- motion field
- real time
- image frames
- object tracking
- motion estimates
- motion estimation
- scene flow
- high frame rate
- motion analysis
- video sequences
- multi camera
- moving camera
- surveillance videos
- smoothness constraint
- image motion
- stereo matching
- event detection
- camera motion
- flow field
- kalman filter
- stereo vision
- stereo images
- motion parameters
- d scene
- particle filter
- depth map
- feature matching
- visual odometry
- temporal order
- moving target
- soccer video
- frame interpolation
- structure from motion
- binocular disparity
- depth data