Motion Estimation Algorithms for Video Compression

Video technology promises to be the key for the transmission of motion video. A number of video compression techniques and standards have been introduced in the past few years, particularly the MPEG-1 and MPEG-2 for interactive multimedia and for digital NTSC and HDTV applications, and H.261/H.263 for video telecommunications. These techniques use motion estimation techniques to reduce the amount of data that is stored and transmitted for each frame. Motion Estimation Algorithms for Video Compression is about these motion estimation algorithms, their complexity, implementations, advantages and drawbacks. First, an overview of video compression techniques is presented with an emphasis on techniques which use motion estimation, such as MPEG and H.261/H.263. Then, the book presents a survey of current motion estimation search algorithms, including the exhaustive search and a number of fast search algorithms. An evaluation of current search algorithms, based on a number of experiments on several test video sequences, is also presented. Motion Estimation Algorithms for Video Compression develops the theoretical framework for a new fast search algorithm, Densely-Centered Uniform-P Search (DCUPS). The complexity of the DCUPS algorithm is comparable to other popular motion estimation techniques; however the algorithm shows superior results in terms of compression ratios and video quality. This book serves as an excellent reference source and may be used as the text for an advanced course on the subject.