Soccernet 2022挑战结果

论文标题

Soccernet 2022挑战结果

SoccerNet 2022 Challenges Results

论文作者

Giancola, Silvio, Cioppa, Anthony, Deliège, Adrien, Magera, Floriane, Somers, Vladimir, Kang, Le, Zhou, Xin, Barnich, Olivier, De Vleeschouwer, Christophe, Alahi, Alexandre, Ghanem, Bernard, Van Droogenbroeck, Marc, Darwish, Abdulrahman, Maglo, Adrien, Clapés, Albert, Luyts, Andreas, Boiarov, Andrei, Xarles, Artur, Orcesi, Astrid, Shah, Avijit, Fan, Baoyu, Comandur, Bharath, Chen, Chen, Zhang, Chen, Zhao, Chen, Lin, Chengzhi, Chan, Cheuk-Yiu, Hui, Chun Chuen, Li, Dengjie, Yang, Fan, Liang, Fan, Da, Fang, Yan, Feng, Yu, Fufu, Wang, Guanshuo, Chan, H. Anthony, Zhu, He, Kan, Hongwei, Chu, Jiaming, Hu, Jianming, Gu, Jianyang, Chen, Jin, Soares, João V. B., Theiner, Jonas, De Corte, Jorge, Brito, José Henrique, Zhang, Jun, Li, Junjie, Liang, Junwei, Shen, Leqi, Ma, Lin, Chen, Lingchi, Marques, Miguel Santos, Azatov, Mike, Kasatkin, Nikita, Wang, Ning, Jia, Qiong, Pham, Quoc Cuong, Ewerth, Ralph, Song, Ran, Li, Rengang, Gade, Rikke, Debien, Ruben, Zhang, Runze, Lee, Sangrok, Escalera, Sergio, Jiang, Shan, Odashima, Shigeyuki, Chen, Shimin, Masui, Shoichi, Ding, Shouhong, Chan, Sin-wai, Chen, Siyu, El-Shabrawy, Tallal, He, Tao, Moeslund, Thomas B., Siu, Wan-Chi, Zhang, Wei, Li, Wei, Wang, Xiangwei, Tan, Xiao, Li, Xiaochuan, Wei, Xiaolin, Ye, Xiaoqing, Liu, Xing, Wang, Xinying, Guo, Yandong, Zhao, Yaqian, Yu, Yi, Li, Yingying, He, Yue, Zhong, Yujie, Guo, Zhenhua, Li, Zhiheng

论文摘要

Soccernet 2022挑战是Soccernet团队组织的第二个年度视频理解挑战。在2022年，挑战由6个基于视觉的任务组成：（1）动作发现，重点是在长期未修饰的视频中检索动作时间戳，（2）重播接地，重点侧重于检索重播中显示的动作的实时时刻，（3）俯仰本地化，集中于检测摄像机和摄像机的摄像机，（4）摄像机量级，（4）摄像机，（4）摄像机，（4）摄像机，（4）摄像机，（4）摄像机，（4）摄像机，（4）摄像机，（4）摄像机的量级，（4）摄像机效果，4）参数，（5）播放器重新识别，重点是从多个视图中检索相同的玩家，以及（6）多个对象跟踪，专注于通过未经编辑的视频流进行跟踪和球。与去年的挑战相比，任务（1-2）重新定义了其评估指标以考虑更紧密的时间准确性，而任务（3-6）是新颖的，包括其基本数据和注释。有关任务，挑战和排行榜的更多信息，请访问https://www.soccer-net.org。基线和开发套件可在https://github.com/soccernet上找到。

The SoccerNet 2022 challenges were the second annual video understanding challenges organized by the SoccerNet team. In 2022, the challenges were composed of 6 vision-based tasks: (1) action spotting, focusing on retrieving action timestamps in long untrimmed videos, (2) replay grounding, focusing on retrieving the live moment of an action shown in a replay, (3) pitch localization, focusing on detecting line and goal part elements, (4) camera calibration, dedicated to retrieving the intrinsic and extrinsic camera parameters, (5) player re-identification, focusing on retrieving the same players across multiple views, and (6) multiple object tracking, focusing on tracking players and the ball through unedited video streams. Compared to last year's challenges, tasks (1-2) had their evaluation metrics redefined to consider tighter temporal accuracies, and tasks (3-6) were novel, including their underlying data and annotations. More information on the tasks, challenges and leaderboards are available on https://www.soccer-net.org. Baselines and development kits are available on https://github.com/SoccerNet.

下载PDF全文

下载文献需遵守相关版权规定

论文标题