The integration of gaze gesture sensors in next-generation smart glasses will improve usability and enable new interaction concepts. However, consumer smart glasses place additional requirements to gaze gesture sensors, such as a low power consumption, high integration capability and robustness to ambient illumination. We propose a novel gaze gesture sensor based on laser feedback interferometry (LFI), which is capable to measure the rotational velocity of the eye as well as the sensors distance towards the eye. This sensor delivers a unique and novel set of features with an outstanding sample rate allowing to not only predict a gaze gesture but also to anticipate it. To take full advantage of the unique sensor features and the high sampling rate, we propose additionally a novel gaze gesture classification algorithm based on single sample. At a mean F1-score of 93.44 performance at a negative latency between gaze gesture input and command execution.
|