You are here: Home / Data and Software

Data and Software


Gaze estimation data base: the EYEDIAP dataset.

The EYEDIAP dataset was designed to train and evaluate gaze estimation algorithms from RGB and RGB-D data. It contains a diversity of participants, head poses, gaze targets and sensing conditions. The recording methodology was designed such that we systematically include, and isolate, most of the variables which affect the remote gaze estimation algorithms:

  • Head pose variations.
  • Person variation.
  • Changes in ambient and sensing condition.
  • Types of target: screen or 3D object.

Some pre-defined benchmarks are provided to evaluate each one of these aspects in an independent manner, and the data was preprocessed to extract and provide complementary observations (e.g. head pose).

Follow theĀ corresponding linkĀ for more information, including download and licence.


RGBD: A Python based RGB-D data processing module.

This python module implements the streaming, calibration and visualization of RGB-D data (combined color and depth images). It generates 3D meshes from the RGB-D camera calibration information and allows its 3D rendering. It is compatible with diverse consumer sensors (Kinect, Primesense, etc...) and recorded data. Python callback functions can be defined to extend the module functionality.

Follow the corresponding link to download.

HG3D: Head Pose and Gaze 3D Tracking Software.

This software implements head pose and gaze estimation algorithms. It builds up on top of the rgbd module, thus allowing the processing of pre-recorded data (such as the EYEDIAP database) or a connected device. It is mostly based on the methodology described in this paper

Follow the corresponding link for more information, including download and licence.