Dataset Name: DLR Dataset (May 2012)
Research Groups: DLR
Hand Type: Human Hand
Data Type: Human Motion Data, Human Postures
Data Structure: Joint Angles (rad)
Data Format: .txt
Sampling Rate: >= 100 Hz
Action Type: Static Grasps, Reach and Grasp
Objects Type: Real Objects
Kin. Model #DOFs: >15 & <= 20 (20)
Equipment: Marker Motion Capture Systems (Vicon)
# of Actions: > 20 (>20 ((6 subjects x 2 tasks) + (1 subject x 1 task)) x 23 objects x m samples per object (time series).)
# of Subjects: > 5 (7)
Year: 2012
Dataset Information:
The objects grasped are a subset of the objects used by Santello et al. [1]. In contrast to [1], here real objects instead of imagined objects are grasped, meaning that the contact forces between the fingers and the object can induce deformations to the hand posture. Each object is associated with a certain use scenario, because this strongly influences the location of the grasp. Seven subjects took part in the measurements; the number of objects is 23.
More information can be found on the report that accompanies the dataset.
How to Cite:
[1] M. Santello, M. Flanders, and J. F. Soechting. Postural hand synergies for tool use. The Journal of Neuroscience, 18(23):10105-10115, 1998.