Real-time monocular depth estimation for low-power embedded systems using deep learning.
Shuyao LiuShuo ZhaoPu ZhangJingjing ChengPublished in: J. Real Time Image Process. (2022)
Keyphrases
- embedded systems
- low power
- low cost
- deep learning
- depth estimation
- real time
- high speed
- depth map
- stereo vision
- stereo matching
- image sequences
- unsupervised learning
- power consumption
- digital camera
- depth information
- dynamic scenes
- weakly supervised
- super resolution
- vision system
- stereo pair
- pattern recognition
- software systems
- signal processing
- real scenes
- three dimensional
- image segmentation