L-EfficientUNet: Lightweight End-to-End Monocular Depth Estimation for Mobile Robots.
Xiangyu LiuJunyan ChenYan ZhouYuexia ZhouJinhai WangXiaoquan OuHaotian LeiPublished in: ICIRA (8) (2023)
Keyphrases
- end to end
- lightweight
- depth estimation
- mobile robot
- monocular images
- depth cues
- stereo vision
- depth map
- depth information
- stereo matching
- dynamic scenes
- depth estimates
- scene understanding
- super resolution
- ad hoc networks
- real scenes
- stereo pair
- image sequences
- disparity map
- feature matching
- pose estimation
- specular reflection
- high quality
- three dimensional
- vision system
- ground truth
- d objects
- stereo images
- object detection
- rfid tags
- d scene
- wireless sensor networks
- wireless networks