Logo GRASP UNDERSTANDING DATASET (GUN-71)

 


 

Description.
We have captured 12,000 RGB-D images of object manipulation scenes (from a chest-mounted RGB-D camera) that we labeled with one of 71 fine-grained grasps defined in a recent taxonomy. We considered 28 objects per grasp, resulting in a total of 1988 different hand-object configurations with 5-6 views for each. We consider 8 different subjects (4 males and 4 females) in 5 different houses. For more details, please see our paper.

Download GUN-71 dataset:

Please cite:

Aknowledgement
This research is supported by the European Commission under FP7-PEOPLE-2012-IOF – Marie Curie Action: “International Outgoing Fellowships for Career Development”. Grant “Egovision4Health”. (PIOF-GA-2012-328288).