Visual Saliency Model using SIFT and Comparison of Learning Approaches

摘要

Humans' ability to detect and locate salient objects on images is remarkablyfast and successful. Performing this process by using eye tracking equipment isexpensive and cannot be easily applied, and computer modeling of this humanbehavior is still a problem to be solved. In our study, one of the largestpublic eye-tracking databases which has fixation points of 15 observers on 1003images is used. In addition to low, medium and high-level features which havebeen used in previous studies, SIFT features extracted from the images are usedto improve the classification accuracy of the models. A second contribution ofthis paper is the comparison and statistical analysis of different machinelearning methods that can be used to train our model. As a result, a bestfeature set and learning model to predict where humans look at images, isdetermined.