A video encoder is disclosed, the encoder for encoding image frames that are divisible into macroblocks. The video encoder includes means for generating a QP estimate for the macroblocks of an image frame. The video encoder further includes means for selection of a frame level quantization parameter (QP) for the image frame, using one of mean, median, and mode of QP estimates for the macroblocks. A method for encoding video is disclosed for encoding image frames that are divisible into macroblocks. The method generates a quantization parameter (QP) estimate for the macroblocks of an image frame and selects a frame level QP for the image frame, using one of mean, median, and mode of QP estimates for the macroblocks.