This work is an early version of our submission paper, "Self-supervised monocular depth estimation for real-world deployment: An economical approach"
We will update the full version after the article is accepted, please consider citing it if you think it is a useful work.
The main framework of this project is based on Monodepth2. You can use the code of this project to replace the corresponding positions of Monodepth2 for training and testing.
- Thank the authors for their superior works: monodepth2, Lite-Mono