LOW-RESOLUTION VIDEO CODING CONTENT EXTRACTION

A method for extracting low-resolution video from compressed high-resolution video is disclosed. First, a video, comprising high-resolution frames encoded with a predictive coding technique, is input. At least one of the high-resolution frames is a reference frame. The reference frame is down-sampled to produce a first low-resolution frame with a first DC coefficient. Frequency coefficients corresponding to the first low-resolution frame are generated. Then, a second low-resolution frame is generated according to the frequency coefficients and the first low-resolution frame, the second low-resolution frame having a second DC coefficient. A residual error corresponding to the second low-resolution frame is generated. Finally, a third low-resolution frame is generated according to the frequency coefficients, the first low-resolution frame, and the residual error. A third DC coefficient for the third low-resolution frame is determined according to a weighted average of at least the first DC coefficient and the second DC coefficient.