Auto-Calibrated Gaze Estimation Using Human Gaze Patterns

Abstract

We present a novel method to auto-calibrate gaze estimators based on gaze patterns obtained from other viewers. Our method is based on the observation that the gaze patterns of humans are indicative of where a new viewer will look at. When a new viewer is looking at a stimulus, we first estimate a topology of gaze points (initial gaze points). Next, these points are transformed so that they match the gaze patterns of other humans to find the correct gaze points. In a flexible uncalibrated setup with a web camera and no chin rest, the proposed method is tested on ten subjects and ten images. The method estimates the gaze points after looking at a stimulus for a few seconds with an average error below 4.5 • . Although the reported performance is lower than what could be achieved with dedicated hardware or calibrated setup, the proposed method still provides sufficient accuracy to trace the viewer attention. This is promising considering the fact that auto-calibration is done in a flexible setup , without the use of a chin rest, and based only on a few seconds of gaze initialization data. To the best of our knowledge, this is the first work to use human gaze patterns in order to auto-calibrate gaze estimators.

Implicit Calibration Using Probable Fixation Targets

“…Using the targets instead of the whole saliency maps as it was typically done in previous research [11,16,18] has several important advantages:…”

“…Instead of using algorithmic saliency model it is also possible to use genuine human gazes recorded earlier for the same image as it was done in [18]. However, such a solution is possible only when the same stimulus is presented many times and is not feasible in general case.…”

“…Bayesian network [11], K-closest points and mixture model methods [18] or minimization of the KL-divergence between distributions [4].…”

“…Most experiments published so far recorded a set of gazes while a person was watching static images [4,11,18,21]. Only in [16] the authors used a set of frames from a movie and computed a saliency map for each frame.…”

3D Gaze Estimation for Head-Mounted Eye Tracking System With Auto-Calibration Method

“…Several approaches have been proposed to avoid the person-specific calibration process. Alnajar et al claim that the gaze patterns of several viewers provide important cues for the auto-calibration in [13]. By making use of the topology of pre-recorded gaze pattern, a transformation is computed to map the initial gaze points to match the gaze pattern.…”