Depth Monocular Estimation with Attention-based Encoder-Decoder Network from Single Image.
Xin ZhangRabab AbdelfattahYuqi SongSamuel A. DauchertXiaofeng wangPublished in: CoRR (2022)
Keyphrases
- single image
- d scene
- shape recovery
- light source
- video codec
- scene geometry
- low complexity
- super resolution
- distributed video coding
- multiple images
- image pairs
- shape from shading
- decoding process
- image sequences
- depth map
- indoor scenes
- depth cues
- partially occluded
- high quality
- ego motion
- rate distortion
- motion estimation
- stereo vision
- video coding
- cast shadows
- pose estimation
- image registration
- error control
- object recognition
- fundamental problem in computer vision
- intrinsic image decomposition