Visual saliency: A manifold way of perception

Visual saliency plays an important role in the human visual system HVS since it is indispensable for object detection and recognition. A bottom-up saliency model was proposed, following the manifold characteristic of HVS, previously developed for understanding HVS mechanism. The saliency of a given location of visual field is defined as the power of features responses after the dimensionality reduction with manifold learning for sparse representation of raw input. This saliency definition also explains the reason that HVS can suppress the response of redundant pattern and excite the response of attended pattern. Experiments show that our saliency model produces better predictions of human eye fixations on two dataset in the comparsion of four state-of-the-art methods.