Systems and methods for up-scaling video

In a system for up-scaling a source video having a lower, first resolution to a desired, up-scaled output video having a higher, second resolution, each video being comprised of a plurality of sequentially-ordered frames of images, the resolution of each respective image being defined by its number of pixels, wherein one or more fractal zooming stages for doubling the resolution is applied to a plurality of individual source pixels of each respective frame of the source video to generate a plurality of candidate replacement pixels to represent each of the corresponding plurality of individual source pixels in the desired, up-scaled output video, wherein each pixel is comprised of color components, wherein each fractal zooming stage generates a 2×2 set of candidate replacement pixels in each color component associated with a corresponding one of the plurality of individual source pixels, the candidate replacement pixels being used to create a proposed replacement frame having, after completion of the one or more fractal zooming stages, an interim resolution, wherein if the interim resolution of the plurality of proposed replacement frames is greater than the second resolution, then the plurality of proposed replacement frames are downsized to a plurality of final frames having the second resolution that, when viewed in sequential order, comprises the desired, up-scaled output video, a method comprising:

• during each fractal zooming stage, for each individual source pixel (A22) and its corresponding 2×2 set of candidate replacement pixels (X00, X01, X10, X11), applying a first artifact filter to each color component to improve the quality of the candidate replacement pixels, wherein the first artifact filter comprises the steps of:
• generating a horizontal correction factor eh as a function of the pixel values of the pixel to the left (A21) and right (A23) of the individual source pixel and the average pixel values of the two leftmost candidate replacement pixels (X00, X10) and of the two rightmost candidate replacement pixels (X01, X11);
• generating a vertical correction factor ev as a function of the pixel values of the pixel above (A12) and below (A32) the individual source pixel and the average pixel values of the two uppermost candidate replacement pixels (X00, X01) and of the two lowermost candidate replacement pixels (X10, X11);
• adding the horizontal correction factor eh to the two leftmost candidate replacement pixels (X00, X10) and subtracting the horizontal correction factor eh from the two rightmost candidate replacement pixels (X01, X11);
• and
• adding the vertical correction factor ev to the two uppermost candidate replacement pixels (X00, X01) and subtracting the vertical correction factor ev from the two lowermost candidate replacement pixels (X10, X11);
• wherein the resulting values of the 2×2 set of candidate replacement pixels are updated to be, respectively, X00+eh+ev, X01−eh+ev, X10+eh−ev, and X11−eh−ev.