Advertisement Evaluation Based On Visual Attention Mechanism
Xiao Yu, Peng Gan, Yuling Sun · 2017
Evaluation methods of advertisement design are important in Advertising, This paper proposed a novel approach to advertisement evaluation based on visual attention.Firstly saliency detection algorithm is used to detect the salient region of advertisement, then an advertisement is determined as attractive if the detected salient regions are overlapped with the interested regions of the advertisement.Based on the salient regions, the fixations are generated in sequence combined with the competition mechanism of WTA and inhibition of return mechanism.The fixations reflects the intensity of each salient region in the advertisement, which determines the priority of the visual effect on each visual element.The experimental result shows that this evaluation method can basically simulate the attention mechanism of human visual system to evaluate the advertisement objectively.