Learning to Fuse Monocular and Multi-view Cues for Multi-frame Depth Estimation in Dynamic Scenes.
Rui LiDong GongWei YinHao ChenYu ZhuKaixuan WangXiaozhi ChenJinqiu SunYanning ZhangPublished in: CVPR (2023)
Keyphrases
- dynamic scenes
- multi view
- depth estimation
- depth map
- multiple views
- single view
- three dimensional
- multiple cameras
- multi frame
- moving camera
- d objects
- ego motion
- image sequences
- motion segmentation
- d scene
- camera calibration
- semi supervised
- video sequences
- space time
- stereo matching
- moving objects
- foreground objects
- range images
- pose estimation
- motion estimation
- reinforcement learning
- background subtraction
- depth information
- viewpoint
- high quality
- stereo correspondence