Monocular depth estimation with hierarchical fusion of dilated CNNs and soft-weighted-sum inference.
Bo LiYuchao DaiMingyi HePublished in: Pattern Recognit. (2018)
Keyphrases
- weighted sum
- depth estimation
- monocular images
- depth cues
- stereo vision
- depth map
- stereo matching
- depth information
- linear combination
- depth estimates
- real scenes
- scene understanding
- objective function
- dynamic scenes
- super resolution
- bayesian networks
- stereo pair
- feature matching
- image sequences
- multi view
- stereo images
- pose estimation
- high quality
- d scene
- structure from motion
- multi sensor
- post processing
- low resolution
- input data
- motion estimation
- computational complexity
- three dimensional