This paper discusses video compression and decompression strategies for use on general purpose computer systems where no specialized hardware is available. We first examine the alternatives and describe the performance and limitations of the first generation of such methods. A brief description is given of the possible algorithmic approaches. We introduce requirements for the encoder/decoder, and show that a vector quantization based scheme with image preprocessing and classification can provide the required performance. The remainder of the paper deals with the algorithm design aiid the tradeoffs made to meet the goals of realtime decode capability, compression ratio and image quality. Finally, we present results for the optimized algorithm and indicate areas which appear to be most promising for further work. BACKGROUND Codec requirementsThe most basic requirement is that the video be decodable in realtime on currently available personal computer platforms. In order for such video to be interesting, motion should be represented smoothly and the image should be of reasonable size. In our opinion, this translates into a minimum of 12-15 frames per second with an image size of 320x240 (quarter of a staidard Macintosh screen). We targeted this playback performance for the CD-ROM device, which is the likely distribution medium for the compressed video material. The limited CD-ROM bandwidth (as compared to hard disks) adds a significant challenge1. Though double speed CD-ROMs are now available (300 Kbytes/s bandwidth), there is also a large installed base of single speed devices. If we assume an available bandwidth of 150 Kbytes/s and negligible decode time, a compression ratio of 23:1 (0.13 bpp) is needed for 15 f/s playback. Therefore, to achieve real time playback, the algorithm must be designed with two goals in mind --the high compression ratio needed to prevent the limited CD-ROM bandwidth from becoming a bottleneck, and the low decode complexity needed to minimize decode time. In the next 2-3 years, the CD-ROM bandwidth problem will be alleviated somewhat with the proliferation of quad speed CD drives. Algorithm optionsAlthough the first generation of software codecs {c. 1990] were not capable of meeting the requirements described above, they did provide limited motion video capability on the desktop thr the first time. The most successful of these methods was based on Block Truncation Coding, a method refined by Hoffert et al [1], ftr fast encoding and realtime decode2 on the Macintosh. Hoffert's method, called "Apple Video" in QuickTime, achieved compression ratios in the range of 2-8: 1 for intraframe coding. For higher compression and higher quality, software versions of JPEG also became available. However, JPEG cannot, as yet provide real-time decode capability due to high decoder computational complexity. Similar problems exist with software versions of MPEG and P*64. Therefore, it became clear that the required compression ratio must be achieved without adding much decode complexity. This impli...
Low bit rate image coding at 10 kbit/s and less is a difficult problem and does not appear possible with the current generation of block transform based methods. Current research efforts center around the use of transforms with less objectionable artifacts such as wavelets or model based methods.We examine a method that is transform based but captures specific features of the image to be represented. The transform uses principal component analysis to generate a basis set specific to the particular class of images to be coded. We present results from a transform designed for use in a "talking head" sequence.Significant improvement in reconstructed quality is shown when perceptual weighting is used in generating the basis set. The appendix includes details of computationally efficient methods for deriving the basis set as well as a description of the weighting method.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations鈥揷itations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright 漏 2024 scite LLC. All rights reserved.
Made with 馃挋 for researchers
Part of the Research Solutions Family.