A method for detecting user liveness is provided that includes selecting firstand second imagesfrom a sequence of images. The first and second images are captured underdifferentillumination conditions. The method further includes locating featurecalculation windows incorresponding positions on the first and second images. Each window includes afirst area and asecond area. Moreover, the method includes calculating, by a computing device,a feature valuefor each window position based on pixels, within the windows located at theposition, from thefirst and second images. Furthermore, the method includes calculating afeature vector from thefeature values, calculating a confidence score from the feature vector, anddetermining thesequence of images includes images of a live user when the confidence score isequal to orgreater than the threshold score.