Online Multi-player Tracking in Monocular Soccer Videos
Online Multi-player Tracking in Monocular Soccer Videos
复制标题
DOI：
10.1016/j.aasri.2014.08.006
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
M. Herrmann;M. Hoernig;B. Radig
中科院分区：
文献类型：
--
作者：
M. Herrmann;M. Hoernig;B. Radig
在单目足球视频中跟踪球员是一项具有挑战性的任务，因为可能会发生许多困难，特别是在电视广播中，例如摄像机运动，球员的严重遮挡或不均匀的闪电条件。我们提出了一种新的鲁棒多玩家跟踪方法，这是基于找到局部最大值的置信图。该地图表示视觉证据的集合，例如团队服装的颜色，HOG人类探测器的响应以及图像中的草地区域。这些特征的组合允许鲁棒的在线跟踪过程，其不需要关于相机校准或其他用户输入的任何进一步信息。在使用四个有代表性的数据集的评估中，我们的算法显示出显着的准确性，并优于最先进的行人跟踪器。
The tracking of players in monocular soccer videos is a challenging task because of numerous difficulties that can occur especially in TV broadcasts, such as camera motions, severe occlusion of players, or inhomogeneous lightning conditions. We propose a new robust method for multi-player tracking, which is based on finding local maxima on a confidence map. This map represents an ensemble of visual evidences, such as colors of the team outfits, responses of a HOG human detector, and grass regions in images. This combination of features allows for a robust online tracking procedure that does not require any further information about the camera calibration or other user input. In the evaluation using four representative datasets, our algorithm shows remarkable accuracy and outperforms a state-of-the-art pedestrian tracker.