Figure 1From: Model of visual attention for video sequencesInitial basis images. a) the original image; b) a set of basis functions received by ICA from several patches from one image; c) and d) the convolution results of two functions with the original image.Back to article page