User interface control based on head orientation

A method comprising:

• receiving, by a computing device, coordinates representing a set of at least three reference points in an image of a subject gazing on a plurality of user interface elements, the set including a first reference point and a second reference point located on opposite sides of a third reference point, wherein the subject is viewed by a camera, and wherein receiving the coordinates comprises receiving a height of the subject and a distance of the subject from the camera;
• more of the following:
• a height of the subject and a distance from the computing device of the subject;
• calculating the third reference point based on the height and the distance;
• determining a first distance between the first reference point and the third reference point;
• determining a second distance between the second reference point and the third reference point;
• comparing, by the computing device, the determined first distance to the determined second distance to calculate a head orientation value for the subject, the first distance and the second distance being weighted to calculate the head orientation value;
• and
• selecting, by the computing device, at least one of the plurality of user interface elements based on the calculated head orientation value.

One or more computer storage media embodying computer-executable components, said components comprising:

• a communications interface component that when executed causes at least one processor to receive coordinates representing a first set of at least three reference points from a first video frame of a subject gazing on a plurality of user interface elements, the set including a first reference point and a second reference point located on opposite sides of a third reference point, wherein the subject is viewed by a capture device, and wherein receiving the coordinates comprises receiving a height of the subject and a distance of the subject from the capture device, wherein the third reference point is calculated based on the height and the distance of the subject;
• a detection component that when executed causes at least one processor to determine a first distance between the first reference point and the third reference point and to determine a second distance between the second reference point and the third reference point;
• and
• a score component that when executed causes at least one processor to compare the first distance determined by the detection component to the second distance determined by the detection component to calculate a first head orientation value for the subject, the first distance and the second distance being weighted to calculate the head orientation value,
• wherein the detection component and the score component operate on a second set of at least three reference points from a second video frame to calculate a second head orientation value for the subject;
• and
• an interaction component that when executed causes at least one processor to select at least one of the plurality of user interface elements based on a comparison between the first head orientation value calculated by the score component to the second head orientation value calculated by the score component.