KINE-ADL BE-UJI: Kinematics of human hand dataset for feeding and cooking tasks
This dataset contains a database of human hand kinematics with data collected during the performance of a wide variety of activities of daily living involving feeding and cooking. The data were recorded using CyberGlove instrumented gloves on both hands measuring 18 degrees of freedom on each. A total of 20 subjects participated in the recordings, and performed the tasks in a natural non-directed way using 66 different objects. This dataset contains a total of 1160 continuous calibrated recordings taken at 100Hz during the performance of the tasks, with filtered signal. This database can be useful for machine learning purposes and prostheses control, as well as for the characterization of healthy human hand kinematics.
Steps to reproduce
Described in a pdf file included in the dataset (KINE-ADL BE-UJI DATASET_Guide.pdf)