The Second Monocular Depth Estimation Challenge
Abstract
This paper discusses the results for the second edition of the Monocular Depth Estimation Challenge (MDEC). This edition was open to methods using any form of supervision, including fully-supervised, self-supervised, multi-task or proxy depth. The challenge was based around the SYNS-Patches dataset, which features a wide diversity of environments with high-quality dense ground-truth. This includes complex natural environments, e.g. forests or fields, which are greatly underrepresented in current benchmarks. The challenge received eight unique submissions that outperformed the provided SotA baseline on any of the pointcloud- or image-based metrics. The top supervised submission improved relative F-Score by 27.62%, while the top self-supervised improved it by 16.61%. Supervised submissions generally leveraged large collections of datasets to improve data diversity. Self-supervised submissions instead updated the network architecture and pretrained backbones. These results represent a significant progress in the field, while highlighting avenues for future research, such as reducing interpolation artifacts at depth boundaries, improving self-supervised indoor performance and overall natural image accuracy.
Cite
@article{arxiv.2304.07051,
title = {The Second Monocular Depth Estimation Challenge},
author = {Jaime Spencer and C. Stella Qian and Michaela Trescakova and Chris Russell and Simon Hadfield and Erich W. Graf and Wendy J. Adams and Andrew J. Schofield and James Elder and Richard Bowden and Ali Anwar and Hao Chen and Xiaozhi Chen and Kai Cheng and Yuchao Dai and Huynh Thai Hoa and Sadat Hossain and Jianmian Huang and Mohan Jing and Bo Li and Chao Li and Baojun Li and Zhiwen Liu and Stefano Mattoccia and Siegfried Mercelis and Myungwoo Nam and Matteo Poggi and Xiaohua Qi and Jiahui Ren and Yang Tang and Fabio Tosi and Linh Trinh and S. M. Nadim Uddin and Khan Muhammad Umair and Kaixuan Wang and Yufei Wang and Yixing Wang and Mochu Xiang and Guangkai Xu and Wei Yin and Jun Yu and Qi Zhang and Chaoqiang Zhao},
journal= {arXiv preprint arXiv:2304.07051},
year = {2023}
}
Comments
Published at CVPRW2023