A method includes: performing spatio-temporal analysis on each frame of a video frame sequence; determining, based on the spatio-temporal analysis, how well a first region of a first frame of the video frame sequence meets a human-perception criterion; assigning, based on the identification, a bit weight for an encoder to apply in encoding the first region; and encoding, using the encoder, at least the first frame based on the spatio-temporal analysis, wherein the first region is encoded using the assigned bit weight.