Fixation Smoothing and Saccade Detection Algorithm

Kumar, Klingner et al. – Improving the Accuracy of Gaze
http://portal.acm.org/citation.cfm?id=1344488

This article describes an algorithm which determines by the gaze data whether the user is currently starting a saccade or just a microsaccade. If it is a miccrosaccade the current gaze data is not considered. Thus a better stability during a fixation can be achieved.

The algorithm encounters the problem of eye-noise:

fixations are not stable and the eye jitters during fixations due to drift, tremor and involuntary micro-saccades [Yarbus 1967]. This gaze jitter, together with the limited accuracy of eye trackers, results in a noisy gaze signal

As the analysis of the data is done in real time a minimal lag occurs. One data record is processed and afterwards the mouse pointer is set, or not. This results in a one-data-sample lag.

Error rates with gaze pointing and selection are hight than with mouse:

In the paper describing EyePoint [Kumar et al. 2007b], it was reported that while the speed of a gaze-based pointing technique was comparable to the mouse, error rates were significantly higher.

Relevance of this article

In this paper we present three methods for improving the accuracy and user experience of gaze-based pointing: an algorithm for realtime saccade detection and fixation smoothing, an algorithm for improving eye-hand coordination, and the use of focus points. These methods boost the basic performance for using gaze information in interactive applications and in our applications made the difference between prohibitively high error rates and practical usefulness of gaze-based interaction.

Method of the algorithm (TODO: I NEED TO UNDERSTAND THIS):

To smooth the data from the eye tracker in real-time, it is necessary to determine whether the most recent data point is the beginning of a saccade, a continuation of the current fixation or an outlier relative to the current fixation. We use a gaze movement threshold, in which two gaze points separated by a Euclidean distance of more than a given saccade threshold are labeled as a saccade. This is similar to the velocity threshold technique described in [Salvucci and Goldberg 2000], with two modifications to make it more robust to noise. First, we measure the displacement of each eye movement relative to the current estimate of the fixation location rather than to the previous measurement. Second, we look ahead one measurement and reject movements over the saccade threshold which immediately return to the current fixation.