HUMAN VISUAL PERCEPTION CONCEPTS AS MECHANISMS FOR SALIENCY DETECTION
Oana Loredana Buzatu · 2012
Predicting human eye fixations is a difficult task that usually involves complex knowledge about the human visual per- ception pathways. This paper proposes a method to detect salient locations using a psycho visual space based on color perception, perceptual decomposition of visual information in multiple processing channels and contrast sensitivity as basic concepts. Using a linear support vector machine to train a saliency detector directly from human eye fixation data, salient locations were predicted as a linear combination of features. Testing results have proved that the proposed method performs better than other state-of-the-art methods with certain common concepts.