360Motion-Dataset 运动捕捉数据集
收藏资源简介:
360Motion-Dataset 是一个 360 度运动捕捉数据集,用于研究和开发三维运动和行为分析相关的算法,由 香港大学、快手科技和浙江大学于 2024 年发布,相关论文成果为「3DTrajMaster: Mastering 3D Trajectory for Multi-Entity Motion in Video Generation」。这个数据集的 V1 版本包含了 72,000 个视频,这些视频涵盖了 50 种不同的实体,如各种动物,以及 6 个虚幻引擎 (UE) 场景,包括 1 个沙漠场景和 2 个 HDRI 场景。此外,数据集还包含了 121 个不同的轨迹模板,为研究者提供了丰富的运动模式和行为变化。
360Motion-Dataset is a 360-degree motion capture dataset for research and development of algorithms related to 3D motion and behavior analysis. It was released in 2024 by The University of Hong Kong, Kuaishou Technology and Zhejiang University, with its associated paper titled "3DTrajMaster: Mastering 3D Trajectory for Multi-Entity Motion in Video Generation". The V1 version of this dataset contains 72,000 videos covering 50 distinct entity types such as various animals, as well as 6 Unreal Engine (UE) scenes including 1 desert scene and 2 HDRI scenes. Additionally, the dataset includes 121 different trajectory templates, providing researchers with rich motion patterns and behavioral variations.




