GaitMotion
收藏资源简介:
GaitMotion数据集是由不列颠哥伦比亚大学电气与计算机工程系开发的多任务数据集,专注于病理步态预测。该数据集通过穿戴传感器捕捉患者的实时运动数据,包含详细的地面实况标注,支持多种任务,如步态/步幅分割和步态/步幅长度预测。数据集不仅适用于医疗产品中的患者进展监测和疾病后恢复评估,还可用于法医技术中的人员再识别和仿生学研究,以辅助人形机器人的开发。此外,数据集还考虑了个体间数据分布的漂移,这种漂移可能归因于每个参与者的独特行为习惯或传感器潜在的位移。
The GaitMotion dataset is a multi-task dataset developed by the Department of Electrical and Computer Engineering at the University of British Columbia, focusing on pathological gait prediction. This dataset captures real-time motion data of patients via wearable sensors, and includes detailed ground truth annotations, supporting multiple tasks such as gait/stride segmentation and gait/stride length prediction. Moreover, this dataset is applicable not only to patient progress monitoring and post-disease recovery assessment in medical products, but also to person re-identification in forensic technology and bionics research to aid the development of humanoid robots. In addition, the dataset also accounts for data distribution drift among individuals, which may be attributed to each participant's unique behavioral habits or potential displacement of the sensors.

- 1GaitMotion: A Multitask Dataset for Pathological Gait Forecasting不列颠哥伦比亚大学电气与计算机工程系 · 2024年



