Abstract
Smart cameras are used in a large range of applications. Usually the smart cameras transmit the video or/and extracted information from the video scene, frequently on compressed format to fit with the application requirements. An efficient hardware accelerator that can be adapted and provide the required coding performances according to the events detected in the video, the available network bandwidth or user requirements, is therefore a key element for smart camera solutions. We propose in this paper to focus on a key part of the compression system: motion estimation. We have developed a flexible hardware implementation of the motion estimator based on FPGA component, fully compatible with H.264, which enables the integer motion search, the fractional search and variable block size to be selected and adjusted. The main contributions of this paper are the definition of an architecture allowing flexibility and some new hardware optimizations of the architecture of the motion estimation allowing the improvement of the performances (computing time or hardware resources) compared to the state of the art. The paper describes the design and proposes a comparison with state-of-art architectures. The obtained FPGA based architecture can process integer motion estimation on 720 x 576 video streams at 67 fps using full search strategy, and sub-pel refinement up to 650 KMacroblocks/s. (C) 2013 Elsevier B.V. All rights reserved.