Deep Reinforcement Learning of Volume-Guided Progressive View Inpainting for 3D Point Scene Completion From a Single Depth Image.
Xiaoguang HanZhaoxuan ZhangDong DuMingdai YangJingming YuPan PanXin YangLigang LiuZixiang XiongShuguang CuiPublished in: CVPR (2019)
Keyphrases
- depth images
- reinforcement learning
- virtual view
- depth information
- pose estimation
- view dependent
- three dimensional
- depth cameras
- learning algorithm
- random forests
- planar surfaces
- low resolution
- depth map
- d scene
- single image
- input image
- depth data
- time of flight
- stereo vision
- multiple images
- machine learning
- multiple views
- hand held
- view synthesis
- image correspondences
- high resolution
- video sequences
- decision trees
- rgb d camera