GazeSense is a software application that provides 3D eye tracking by relying on consumer 3D sensors. GazeSense allows users to define virtual 3D objects with respect to the camera and world coordinate systems and measure the attention of people towards the objects.

The 3D eye tracking technology by Eyeware can provide a high-level description of attention, which goes beyond the eye tracking signal. GazeSense can answer the following questions in real-time:

• Is the person looking at the table?

• Is participant A looking at participant B?

• Is the user looking at the robot?

• Where is the user looking on the screen?

GazeSense™ also provides a Service that is a stand-alone command-line application which is capable of non-intrusive, wide track-box, real-time 3D eye, head and attention tracking, based on commodity RGB-Depth cameras. GazeSense™ Service provides a custom python API that allows separate `client` applications to connect to the GazeSense Service and to read the real-time tracking data, as well as controlling the behaviour of the GazeSense™ Service itself. This functionality allows third-party applications to become attention-sensing capable with minimal integration effort.

GazeSense™ is also delivered to other platform for easier integrations, as a plugin. We are currently supporting the ROS (Robot Operating System) where GazeSense provides robotics researchers and developers with the real-time signals on attention towards objects. One would be able to tell if a person interacting with the robot is looking at either of the defined objects or not, or if looking at the robot itself. The attention label (‘what is the person looking at’) is then published as a topic into the ROS framework.