A novel vario-power architecture of motion estimation using a content based subsample algorithm

Motion estimation is a key element in many video compression systems; it tends to dominate computational, and hence power, requirements. With the increasing demand for portable, power-aware, multimedia devices, an architecture that can be flexible in both power consumption and compression quality is required. To meet the requirement, the paper presents a novel power-aware architecture, called the vario-power architecture, for motion estimation. Based on a semi-systolic array with a content-based subsample algorithm, the architecture dynamically disables some processing elements to reduce the power consumption. By performing edge extraction first, a threshold is then set as the criterion of whether to enable/disable processing elements and thus the switching activities of the system can be reduced. As a result, the architecture may operate dynamically at different power consumption modes, according to the remaining capacity of the battery pack, with little quality degradation.