[POSTER] Composite Realism: Effects of Object Knowledge and Mismatched Feature Type on Observer Gaze and Subjective Quality
Alan Dolhasz, Maite Frutos-Pascual, Ian Williams · 2017
We report on the results of the first visual search and rating study (N60) evaluating human gaze when assessing the realism of image composites. The effects of object identity knowledge and mismatched feature type on observers' gaze and subjective realism scores are studied. Gaze metrics used include: fixation count, fixation duration, time and duration of first fixation on target object, as well as area of interest similarity and inter-observer consistency. Monte-Carlo-based techniques are used for analysis of the data obtained. Results indicate that while knowledge of object identity impacts gaze allocation and response times, it leaves subjective realism ratings unchanged. We show that the type of mismatched feature (correlated colour temperature vs exposure) has a significant impact on fixation counts and durations. This study provides a first step to utilising objective gaze metrics to better understand subjective assessment processes and leads towards the development of gaze-inspired compositing methods.