EAGAN: Event-based attention generative adversarial networks for optical flow and depth estimation.
Xiuhong LinChenhui YangXuesheng BianWeiquan LiuCheng WangPublished in: IET Comput. Vis. (2022)
Keyphrases
- depth estimation
- optical flow
- depth map
- stereo vision
- stereo matching
- depth information
- image sequences
- dynamic scenes
- d scene
- model based pose estimation
- feature matching
- multi view
- stereo pair
- motion model
- event detection
- moving camera
- moving objects
- scene understanding
- real scenes
- computer vision
- multiple views
- graphical models
- motion segmentation
- dynamic programming
- spatio temporal
- real time