A Fast Motion Estimation Algorithm for H.264

Motion Estimation (ME) plays an important role in digital video compression, since it can significantly affect the image accuracy and the compression ratio. Also, motion estimation process is a part of the encoder which is high complex and computation-intensive, therefore it hinders real-time applications of H.264 video compression. To reduce computational complexity and maintain the image quality, an integer pixel fast motion estimation algorithm is proposed in this paper. Combining the characters of the UMHexagonS algorithm and the EPZS algorithm, the proposed algorithm selects various searching patterns according to the motion types of the image and introduces a early-termination algorithm. Experimental results demonstrate that the proposed algorithm has less ME time than FS, UMHexagonS and EPZS algorithm with essentially the same PSNR.