Hand gestures are one of the natural forms of communication in human-robot interaction scenarios. They can be used to delegate tasks from a human to a robot. To facilitate human-like interaction with robots, a major requirement for advancing in this direction is the availability of a hand gesture dataset for judging the performance of the algorithms.

Dataset Features

• 22 participants performed 8 hand gestures in the context of human-robot interaction scenarios taking place in a close proximity.
• 8 hand gestures categorized as:
  1. 2 types of referencing (pointing) gestures with the ground truth location of the target pointed at,
  2. 2 symbolic gestures,
  3. 2 manipulative gestures,
  4. 2 interactional gestures.
• A corpus of 836 test scenarios (704 referencing gestures with ground truth, and 132 other gestures).
• Hand gestures recorded from two views (frontal and side) using RGB-D Kinect sensor.
• The data acquisition setup can be easily recreated using a polar coordinate pattern as shown in the figure below to add new hand gestures in the future.

• Soon to be released publicly.
• Currently available for Download with authentication.

Sample Scenarios

Gestures recorded from frontal and side view. T-B: Finger pointing, Tool pointing, Thumb up (approve), Thumb down (disapprove), Grasp open, Grasp close, Receive, Fist (stop).

Reference

Dadhichi Shukla, Ozgur Erkent, Justus Piater, The IMHG dataset: A Multi-View Hand Gesture RGB-D Dataset for Human-Robot Interaction. Towards Standardized Experiments in Human Robot Interactions, 2015 (Workshop at IROS). Extended Abstract.PDF.

BibTex

  @InProceedings{Shukla-2015-StandardHRI,
    title = {{The IMHG dataset: A Multi-View Hand Gesture RGB-D Dataset for Human-Robot Interaction}},
    author = {Shukla, Dadhichi and Erkent, Ozgur and Piater, Justus},
    booktitle = {{Towards Standardized Experiments in Human Robot Interactions}},
    year = 2015,
    month = 10,
    note = {Workshop at IROS},
    url = {https://iis.uibk.ac.at/public/papers/Shukla-2015-StandardHRI.pdf}
  }

Acknowledgement

The research leading to these results has received funding from the European Community’s Seventh Framework Programme FP7/2007-2013 (Specific Programme Cooperation, Theme 3, Information and Communication Technologies) under grant agreement no. 610878, 3rd HAND.

Contact

dadhichi[dot]shukla[at]uibk[dot]ac[dot]at