Cue normalization schemes in saliencybased visual attention models

Saliency-based visual attention models provide visual saliency by combining the conspicuity maps relative to var-ious visual cues. Because the cues are of different nature, the maps to be combined show distinct dynamic ranges and a normalization scheme is therefore required. The normal-ization scheme used traditionally is an instantaneous peak-to-peak normalization. It appears however that this scheme performs poorly in cases where the relative contribution of the cues varies significantly, for instance when the kind of scene changes, like when the scene under study becomes un-saturated or worse, when it looses any chromaticity. To rem-edy this drawback, this paper proposes an alternative nor-malization scheme that scales each conspicuity map with re-spect to a long-term estimate of its maximum, a value which is learned initially from a large number of images. The ad-vantage of the new method is first illustrated by several ex-amples where both normalization schemes are compared. Then, the paper presents the results of an evaluation where the computed visual saliency of a set of 40 images is com-pared to the respective human attention as derived from the eye movements by a population of 20 subjects. The better performance of the new normalization scheme demonstrates its capability to deal with scenes of varying type, where cue contributions vary a lot. The proposed scheme seems thus preferable in any general purpose model of visual attention. 1