This is the code to generate the corresponding ground truth depth map of the images in Robotcar dataset,
and the main doc is "eval_oxford.py".
This test sequence "test_files" follows the paper:
@inproceedings{vankadari2020unsupervised,
title={Unsupervised monocular depth estimation for night-time images using adversarial domain feature adaptation},
author={Vankadari, Madhu and Garg, Sourav and Majumder, Anima and Kumar, Swagat and Behera, Ardhendu},
booktitle={European Conference on Computer Vision},
pages={443--459},
year={2020},
organization={Springer}
}
@article{zhao2022unsupervised,
title={Unsupervised monocular depth estimation in highly complex environments},
author={Zhao, Chaoqiang and Tang, Yang and Sun, Qiyu},
journal={IEEE Transactions on Emerging Topics in Computational Intelligence},
volume={6},
number={5},
pages={1237--1246},
year={2022},
publisher={IEEE}
}
The basic code and test sequence are provided by the authors of "Unsupervised monocular depth estimation for night-time images using adversarial domain feature adaptation".