Encoder-Decoder Structure with Multiscale Receptive Field Block for Unsupervised Depth Estimation from Monocular Video.
Songnan ChenJunyu HanMengxia TangRuifang DongJiangming KanPublished in: Remote. Sens. (2022)
Keyphrases
- receptive fields
- multiscale
- depth estimation
- dynamic scenes
- image representation
- space time
- natural images
- stereo vision
- depth map
- pose estimation
- video data
- biologically inspired
- semi supervised
- video sequences
- stereo matching
- video surveillance
- depth information
- image processing
- scale space
- real time
- input data
- d objects
- computational complexity
- computer vision
- stereo pair
- data sets