Fast encoding loss metric

A tangible, non-transitory, machine-readable medium storing instructions that when executed by one or more processors effectuate operations comprising:

• obtaining, with one or more processors, source video data, the source video data having a plurality of frames, the frames each having a plurality of pixel values;
• obtaining, with one or more processors, destination video data, the destination video data being a transformed version of the source video data and having transformed versions of the plurality of frames;
• for a given frame, accessing, with one or more processors, a given source frame among the plurality of frames in the source video data and a given destination frame among the plurality of frames in the destination video data, the given destination frame being a transformed version of the given source frame;
• segmenting, with one or more processors, the given source frame and the given destination frame into a plurality of blocks, each block corresponding to a region of pixels in the respective frame;
• for a plurality of blocks in the given source frame, determining, with one or more processors, a respective source aggregate value that is based on a measure of central tendency of pixel values in the respective block of the given source frame;
• for a plurality of blocks in the given destination frame, determining, with one or more processors, a respective destination aggregate value that is based on a measure of central tendency of pixel values in the respective block of the given destination frame;
• determining, with one or more processors, a plurality of differences between source aggregate values and corresponding destination aggregate values of the given frame;
• determining, with one or more processors, a frame aggregate value that is based on a measure of central tendency of the determined differences between source aggregate values and corresponding destination aggregate values of the given frame;
• determining, with one or more processors, a measure of distortion of the destination video data relative to the source video data based on the frame aggregate value and a plurality of other frame aggregate values of a plurality of other frames;
• and
• storing, with one or more processors, the measure of distortion in memory in association with the destination video data.