BlanketSet
收藏资源简介:
BlanketSet是由生物医学工程研究中心创建的一个临床真实世界数据集,专注于床上的动作识别和半同步定性运动捕捉。该数据集包含405个RGB-IR-D视频记录,涉及14名参与者执行8种不同的运动序列。每个记录重复3次,覆盖不同程度的毯子遮挡,适用于动作识别和人体运动分析系统的定性评估。数据集的创建过程考虑了多种变量,如毯子的初始位置和不同照明条件,以确保数据的多样性和实用性。BlanketSet主要应用于癫痫监测和睡眠分析,旨在解决临床环境中的人体运动分析问题。
BlanketSet is a real-world clinical dataset created by the Biomedical Engineering Research Center, focusing on in-bed motion recognition and semi-synchronous qualitative motion capture. This dataset contains 405 RGB-IR-D video recordings, involving 14 participants performing 8 distinct motion sequences. Each recording was repeated three times, covering varying degrees of blanket occlusion, which is suitable for the qualitative evaluation of motion recognition and human motion analysis systems. The dataset's creation process considered multiple variables such as the initial position of the blanket and different lighting conditions to ensure data diversity and practicality. Primarily applied in epilepsy monitoring and sleep analysis, BlanketSet aims to address the challenges of human motion analysis in clinical settings.

- 1BlanketSet -- A clinical real-world in-bed action recognition and qualitative semi-synchronised MoCap dataset生物医学工程研究中心 · 2023年



