Figure 2. Illustration of the sampling method.

On each trial, we randomly generated a matrix of dimensions 256 × 40 (representing respectively SFs and frames) in which most elements were zeros and a few were ones. We then convolved this sparse matrix with a 2D Gaussian kernel (a “bubble”). This resulted in the trial’s sampling matrix, shown here as a plane with a number of randomly located bubbles. Every column of this sampling matrix was then rotated around its origin to create isotropic 2D random filters. Finally, these 2D random filters were dot-multiplied by the base image’s spectrum and inverse fast Fourier transformed to create a filtered version of the image for every video frame.