Ghost-Depth: A Lightweight Encoder-Decoder Network for Monocular Depth Estimation.
Wei QuanJingjing LiuChao XuCheng HanJia SongShanshan QiPublished in: CNIOT (2023)
Keyphrases
- lightweight
- depth estimation
- monocular images
- depth map
- depth cues
- stereo vision
- depth information
- depth perception
- communication infrastructure
- stereo matching
- wireless sensor networks
- depth estimates
- real scenes
- dynamic scenes
- image sequences
- view synthesis
- distributed video coding
- motion estimation
- stereo pair
- scene understanding
- multi view
- dense depth maps
- d scene
- super resolution
- feature matching
- disparity map
- free viewpoint
- three dimensional
- epipolar geometry
- bit rate
- single image
- high resolution
- computational complexity
- video sequences
- high quality
- rate distortion