Dense Continuous-Time Optical Flow from Events and Frames.
Mathias GehrigManasi MuglikarDavide ScaramuzzaPublished in: CoRR (2022)
Keyphrases
- optical flow
- dense optical flow
- consecutive frames
- moving objects
- scene flow
- high frame rate
- optical flow fields
- optical flow computation
- multi frame
- motion field estimation
- single frame
- dense correspondence
- optical flow field
- motion features
- motion model
- flow estimation
- frame interpolation
- optical flow estimation
- image sequences
- motion field
- flow field
- interesting events
- motion estimation
- event detection
- least squares
- camera motion
- moving objects in video sequences
- dense motion estimation
- velocity field
- video frames
- adjacent frames
- vector field
- motion estimates
- point tracking
- static scene
- point to point correspondences
- motion segmentation
- computer vision
- spatio temporal
- feature tracking
- dynamical systems
- qualitative and quantitative evaluation
- temporal information
- video sequences
- image frames
- structure from motion
- d scene
- motion boundaries
- image pairs