360Motion-Dataset Motion Capture Dataset

360Motion-Dataset is a 360-degree motion capture dataset used to study and develop algorithms related to 3D motion and behavior analysis. It was released by the University of Hong Kong, Kuaishou Technology, and Zhejiang University in 2024. The related paper results are "3DTrajMaster: Mastering 3D Trajectory for Multi-Entity Motion in Video Generation". The V1 version of this dataset contains 72,000 videos covering 50 different entities, such as various animals, and 6 Unreal Engine (UE) scenes, including 1 desert scene and 2 HDRI scenes. In addition, the dataset also contains 121 different trajectory templates, providing researchers with rich movement patterns and behavior changes.

The dataset has two video resolutions, 480×720 and 384×672, each video contains 99 frames, lasts 3.3 seconds, and is captured at a frame rate of 30 frames per second. These videos not only record the movement of the entity, but also contain detailed metadata, such as position data and HDRI information, which are stored in JSON files. Role information and transformation information are also recorded in detail, providing additional context for analysis.