EDFIDepth: enriched multi-path vision transformer feature interaction networks for monocular depth estimation.
Chenxing XiaMengge ZhangXiuju GaoBin GeKuan-Ching LiXianjin FangYan ZhangXingzhu LiangPublished in: J. Supercomput. (2024)
Keyphrases
- multipath
- depth estimation
- interaction networks
- stereo vision
- depth map
- end to end
- stereo matching
- vision system
- computer vision
- dynamic scenes
- image sequences
- routing algorithm
- real scenes
- scene understanding
- stereo pair
- super resolution
- feature vectors
- real time
- depth information
- disparity map
- pose estimation
- d objects
- high quality
- dynamic programming
- three dimensional
- image processing
- video sequences