volumetric video viewing behavior dataset
收藏资源简介:
本数据集由香港中文大学(深圳)未来网络智能研究所发布,是首个大规模的体积视频观看行为数据集,包含50名用户的详细交互数据,涵盖多种场景和条件。数据集内容包括用户的视口、凝视和运动特征,旨在通过深入分析用户行为,优化体积视频的流媒体传输。创建过程中,研究人员从现有的公开体积视频数据集中选取视频,通过Meta Quest Pro头戴设备收集用户数据。该数据集的应用领域包括体积视频流优化、用户行为预测等,旨在解决体积视频传输中的带宽限制问题。
This dataset is released by the Research Institute of Future Network Intelligence, The Chinese University of Hong Kong, Shenzhen. It is the first large-scale dataset for volumetric video viewing behavior, containing detailed interaction data from 50 users across diverse scenarios and conditions. The dataset covers users' viewport, gaze and motion characteristics, aiming to optimize the streaming transmission of volumetric video through in-depth analysis of user behavior. During its creation, researchers selected videos from existing public volumetric video datasets and collected user data using Meta Quest Pro head-mounted displays. Its application scenarios include volumetric video streaming optimization, user behavior prediction and others, aiming to address the bandwidth limitation issues in volumetric video transmission.




