Fast motion vector and bitrate re-estimation for arbitrary downsizing video transcoding

Abstract

In video transcoding technique, many fast methods have been proposed to estimate the motion vectors required for downsizing a pre-coded video by an integer factor. In this paper we examine and compare several fast video transcoding methods for downsizing a pre-coded video by an arbitrary factor. Methods which out-perform others are identified and discussed. We also propose a new method to model the relation between bitrate and frame size under low bitrate condition. With this new model, we can re-estimate the bitrate according to a required frame size or select a suitable frame size for a given target bitrate, while maintaining good video quality. Experimental results are presented to show the performance of the proposed fast motion vector and bitrate re-estimation methods.