A computer-implemented method for method for detecting features in an image, comprising, receiving a first image at a GPU, wherein the GPU comprises a plurality of memory units and wherein the first image is stored in a first memory unit of the plurality of memory units, processing a second image stored in a second memory unit, of the plurality of memory units, to detect one or more features within the second image and writing one or more processed portions of the second image to a third memory unit of the plurality of memory units. In certain aspects, the method further comprises steps for outputting a third image stored in a fourth memory unit of the plurality of memory units. A mobile computing device and GPU are also provided.