A loose-strategy-based Complexity Scalable Motion Estimation method for video coding

In this paper, a new Complexity Scalable Motion Estimation (CSME) method is proposed which can perform Motion Estimation adaptively under different computation or power budgets while keeping high coding performance. We first propose to use a new class-based method to measure the Macroblock (MB) importance. Based on the new MB importance measure, a complete CSME framework is then proposed. The proposed method allocates computation in a one-pass flow and is based on a loose strategy where frames can borrow computation from other frames to achieve better performance under the same computation budget. Experimental results demonstrate that the proposed method can allocate computation more accurately than previous methods.