The Third Monocular Depth Estimation Challenge

• Published in: arXiv.org
• Main Authors: Spencer, Jaime, Tosi, Fabio, Poggi, Matteo, Ripudaman Singh Arora, Russell, Chris, Hadfield, Simon, Bowden, Richard, Zhou, GuangYuan, Li, ZhengXin, Rao, Qiang, Bao, YiPing, Liu, Xiao, Kim, Dohyeong, Kim, Jinseong, Kim, Myunghyun, Lavreniuk, Mykola, Li, Rui, Mao, Qing, Wu, Jiang, Zhu, Yu, Sun, Jinqiu, Zhang, Yanning, Patni, Suraj, Agarwal, Aradhye, Arora, Chetan, Sun, Pihai, Jiang, Kui, Wu, Gang, Liu, Jian, Liu, Xianming, Jiang, Junjun, Zhang, Xidan, Wei, Jianing, Wang, Fangjun, Tan, Zhiming, Wang, Jiabao, Luginov, Albert, Shahzad, Muhammad, Hosseini, Seyed, Trajcevski, Aleksander, Elder, James H
• Format: Paper
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 27.04.2024
• ISSN: 2331-8422

This paper discusses the results of the third edition of the Monocular Depth Estimation Challenge (MDEC). The challenge focuses on zero-shot generalization to the challenging SYNS-Patches dataset, featuring complex scenes in natural and indoor settings. As with the previous edition, methods can use any form of supervision, i.e. supervised or self-supervised. The challenge received a total of 19 submissions outperforming the baseline on the test set: 10 among them submitted a report describing their approach, highlighting a diffused use of foundational models such as Depth Anything at the core of their method. The challenge winners drastically improved 3D F-Score performance, from 17.51% to 23.72%.