학술논문
1st Workshop on Maritime Computer Vision (MaCVi) 2023: Challenge Results
Document Type
Conference
Author
Kiefer, Benjamin; Kristan, Matej; Pers, Janez; Zust, Lojze; Poiesi, Fabio; De Alcantara Andrade, Fabio Augusto; Bernardino, Alexandre; Dawkins, Matthew; Raitoharju, Jenni; Quan, Yitong; Atmaca, Adem; Hofer, Timon; Zhang, Qiming; Xu, Yufei; Zhang, Jing; Tao, Dacheng; Sommer, Lars; Spraul, Raphael; Zhao, Hangyue; Zhang, Hongpu; Zhao, Yanyun; Augustin, Jan Lukas; Jeon, Eui-Ik; Lee, Impyeong; Zedda, Luca; Loddo, Andrea; Di Ruberto, Cecilia; Verma, Sagar; Gupta, Siddharth; Muralidhara, Shishir; Hegde, Niharika; Xing, Daitao; Evangeliou, Nikolaos; Tzes, Anthony; Bartl, Vojtech; Spanhel, Jakub; Herout, Adam; Bhowmik, Neelanjan; Breckon, Toby P.; Kundargi, Shivanand; Anvekar, Tejas; Tabib, Ramesh Ashok; Mudengudi, Uma; Vats, Arpita; Song, Yang; Liu, Delong; Li, Yonglin; Li, Shuman; Tan, Chenhao; Lan, Long; Somers, Vladimir; De Vleeschouwer, Christophe; Alahi, Alexandre; Huang, Hsiang-Wei; Yang, Cheng-Yen; Hwang, Jenq-Neng; Kim, Pyong-Kun; Kim, Kwangju; Lee, Kyoungoh; Jiang, Shuai; Li, Haiwen; Ziqiang, Zheng; Vu, Tuan-Anh; Nguyen-Truong, Hai; Yeung, Sai-Kit; Jia, Zhuang; Yang, Sophia; Hsu, Chih-Chung; Hou, Xiu-Yu; Jhang, Yu-An; Yang, Simon; Yang, Mau-Tsuen
Source
2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) WACVW Applications of Computer Vision Workshops (WACVW), 2023 IEEE/CVF Winter Conference on. :265-302 Jan, 2023
Subject
Language
ISSN
2690-621X
Abstract
The 1 st Workshop on Maritime Computer Vision (MaCVi) 2023 focused on maritime computer vision for Unmanned Aerial Vehicles (UAV) and Unmanned Surface Vehicle (USV), and organized several subchallenges in this domain: (i) UAV-based Maritime Object Detection, (ii) UAV-based Mar-itime Object Tracking, (iii) USV-based Maritime Obstacle Segmentation and (iv) USV-based Maritime Obstacle Detection. The subchallenges were based on the SeaDronesSee and MODS benchmarks. This report summarizes the main findings of the individual subchallenges and introduces a new benchmark, called SeaDronesSee Object Detection v2, which extends the previous benchmark by including more classes and footage. We provide statistical and qualitative analyses, and assess trends in the best-performing method-ologies of over 130 submissions. The methods are sum-marized in the appendix. The datasets, evaluation code and the leaderboard are publicly available (https://seadronessee.cs.uni-tuebingen.de/macvi).