Adaptive Compression Coding

Following an exchange of correspondence, I met Ross in Adelaide in June 1988. I was approached by the University of Adelaide about being an external examiner for this dissertation and willingly agreed. Upon receiving a copy of this work, what struck me most was the scholarship with which Ross approaches and advances this relatively new field of adaptive data compression. This scholarship, coupled with the ability to express himself clearly using figures, tables, and incisive prose, demanded that Ross's dissertation be given a wider audience. And so this thesis was brought to the attention of Kluwer. The modern data compression paradigm furthered by this work is based upon the separation of adaptive context modelling, adaptive statistics, and arithmetic coding. This work offers the most complete bibliography on this subject I am aware of. It provides an excellent and lucid review of the field, and should be equally as beneficial to newcomers as to those of us already in the field.

Real-Time Video Compression: Techniques and Algorithms introduces the XYZ video compression technique, which operates in three dimensions, eliminating the overhead of motion estimation. First, video compression standards, MPEG and H.261/H.263, are described. They both use asymmetric compression algorithms, based on motion estimation. Their encoders are much more complex than decoders. The XYZ technique uses a symmetric algorithm, based on the Three-Dimensional Discrete Cosine Transform (3D-DCT). 3D-DCT was originally suggested for compression about twenty years ago; however, at that time the computational complexity of the algorithm was too high, it required large buffer memory, and was not as effective as motion estimation. We have resurrected the 3D-DCT-based video compression algorithm by developing several enhancements to the original algorithm. These enhancements make the algorithm feasible for real-time video compression in applications such as video-on-demand, interactive multimedia, and videoconferencing. The demonstrated results, presented in this book, suggest that the XYZ video compression technique is not only a fast algorithm, but also provides superior compression ratios and high quality of the video compared to existing standard techniques, such as MPEG and H.261/H.263. The elegance of the XYZ technique is in its simplicity, which leads to inexpensive VLSI implementation of any XYZ codec. Real-Time Video Compression: Techniques and Algorithms can be used as a text for graduate students and researchers working in the area of real-time video compression. In addition, the book serves as an essential reference for professionals in the field.

Recursive Block Coding, a new image data compression technique that has its roots in noncausal models for 1d and 2d signals, is the subject of this book. The underlying theory provides a multitude of compression algorithms that encompass two course coding, quad tree coding, hybrid coding and so on. Since the noncausal models provide a fundamentally different image representation, they lead to new approaches to many existing algorithms, including useful approaches for asymmetric, progressive, and adaptive coding techniques. On the theoretical front, the basic result shows that a random field (an ensemble of images) can be coded block by block such that the interblock redundancy can be completely removed while the individual blocks are transform coded. On the practical side, the artifact of tiling, a block boundary effect, present in conventional block by block transform coding techniques has been greatly suppressed. This book contains not only a theoretical discussion of the algorithms but also exhaustive simulation and suggested methodologies for ensemble design techniques. Each of the resulting algorithms has been applied to twelve images over a wide range of image data rates and the results are reported using subjective descriptions, photographs, mathematical MSE values, and h-plots, a recently proposed graphical representation showing a high level of agreement with image quality as judged subjectively.

Described by Jeff Prosise of PC Magazine as one of my favorite books on applied computer technology, this updated second edition brings you fully up-to-date on the latest developments in the data compression field. It thoroughly covers the various data compression techniques including compression of binary programs, data, sound, and graphics. Each technique is illustrated with a completely functional C program that demonstrates how data compression works and how it can be readily incorporated into your own compression programs. The accompanying disk contains the code files that demonstrate the various techniques of data compression found in the book.

James A. Storer Computer Science Dept. Brandeis University Waltham, MA 02254 Data compression is the process of encoding a body of data to reduce stor age requirements. With Lossless compression, data can be decompressed to be identical to the original, whereas with lossy compression, decompressed data may be an acceptable approximation (according to some fidelity criterion) to the original. For example, with digitized video, it may only be necessary that the decompressed video look as good as the original to the human eye. The two primary functions of data compression are: Storage: The capacity of a storage device can be effectively increased with data compression software or hardware that compresses a body of data on its way to the storage device and decompress it when it is retrieved. Communications: The bandwidth of a digital communication link can be effectively increased by compressing data at the sending end and decom pressing data at the receiving end. Here it can be crucial that compression and decompression can be performed in real time.