MMAIndoor: Patched MLP and multi-dimensional cross attention based self-supervised indoor depth estimation.
Chen LvChenggong HanTianshu SongHe JiangQiqi KouJiansheng QianDeqiang ChengPublished in: Neurocomputing (2024)
Keyphrases
- depth estimation
- multi dimensional
- depth map
- stereo vision
- stereo matching
- scene understanding
- dynamic scenes
- real scenes
- super resolution
- stereo pair
- depth information
- feature matching
- high quality
- data sets
- model based pose estimation
- multi view
- augmented reality
- d scene
- stereo images
- vision system
- motion estimation
- dynamic programming
- spatio temporal
- object recognition
- computer vision