Dexter 1: A Dataset for Evaluation of 3D Articulated Hand Motion Tracking
================================================================================

Dexter 1 is a dataset for evaluating algorithms for markerless, 3D articulated
hand motion tracking. This dataset consists of 7 sequences of general hand
motion that covers the abduction-adduction and flexion-extension ranges of the
hand. All sequences are with a single actor's right hand.

Dexter 1 consists of data from the following sensors.
- 5 Sony DFW-V500 RGB cameras
- 1 Creative Gesture Camera (Close-range ToF depth sensor)
- 1 Kinect structured light camera

While the RGB and ToF data is nearly synchronized, the structured light data is
not synchronized.

In addition, this dataset also contains manual annotations on the ToF data for
all fingertip positions and approximate palm center.


--------------------
Download
--------------------
Compressed Tarball:
http://resources.mpi-inf.mpg.de/handtracker/data/dexter1.tar.gz

Browse:
http://resources.mpi-inf.mpg.de/handtracker/data/dexter1

Camera Setup:
http://resources.mpi-inf.mpg.de/handtracker/data/dexter1/preview/rgb_cameras.jpg
http://resources.mpi-inf.mpg.de/handtracker/data/dexter1/preview/depth_cams.jpg

Note that the RGB data used in the paper and previews has been enhanced by
adjusting color levels. See script AdjustLevels.sh for more details.


--------------------
Sequence Details
--------------------
- adbadd: Abduction-adduction of all fingers together. Please note the spelling.
- fingercount: Counting using each finger and the thumb.
- fingerwave: Waving of fingers.
- flexex1: Flexion-extension of all fingers.
- pinch: Pinching while moving around.
- tigergrasp: Making a posture like a tiger grasping.
- random: Random motions with articulation.


--------------------
Dataset Structure
--------------------
The root directory (containing this file) consists of 3 sub-directories.
- calibration: Contains information for calibrating the cameras for intrinsic
  	       and extrinsic parameters. We recomment using the MATLAB
  	       calibration toolbox.
- preview: Montage video preview of 3 cameras and structured light data for all
  	   sequences.
- data: All the data resides here.
  - SEQUENCE_NAME
    - multicam: Data from RGB cameras stored as PNG files in separate
      		directories.
    - tof: The ToF data.
      - confmap: Confidence map as 16-bit PNG.
      - depth: Depth map as 16-bit PNG.
      - rgb: Empty.
      - uvmap: Emtpy.
      - vertices: Point cloud from depth map stored in PCD format (see
      		  pointclouds.org). Units are mm.
    - structlight: Unsynchronized structured light data in ONI format.
    - annotations: Contains manually annotated data and a preview of
      		   annotations. See README.txt inside the directory for more
      		   details.


--------------------
Citation
--------------------
If you use this dataset in your work, please cite the following paper.

[1] S. Sridhar, A. Oulasvirta, and C. Theobalt, Interactive Markerless
Articulated Hand Motion Tracking Using RGB and Depth Data, Proceedings of the
IEEE International Conference on Computer Vision (ICCV), Sydney, Australia,
2013.

For questions about the dataset please contact Srinath Sridhar
(ssridhar@mpi-inf.mpg.de or srinaths@umich.edu).


--------------------
Acknowledgments
--------------------
We thank Thomas Helten for helping with data capture.