An embedded wavelet hierarchical image coder

Abstract

A Subsidiary of SRI International CN5300 Princeton, NJ 08543-5300. This paper describes a simple, yet remarkably effective, image compression algorithm, having the property that the bits in the bit stream are generated in order of importance, yielding fully hierarchical image compression suitable for embedded coding or progressive transmission. Given an image bit stream, the decoder can cease decoding at any point in the bit stream and still produce exactly the same image that would have been encoded at the bit rate corresponding to the truncated bit stream. The compression algorithm is based on three key concepts: 1) wavelet transform or hierarchical subband decomposition, 2) prediction of the absence of significant information across scales by exploiting the self-similarity inherent in images, and 3) hierarchical entropy-coded quantization.

Directional Transforms for Video Coding Based on Lifting on Graphs

“…This can be extended to our system so that subpixel motion estimation is invertible using samples interpolated (in the spatial or temporal domains) from U nodes. While the proposed entropy coder is similar to CAVLC, it could be improved exploiting intersubband correlations [51], [52] in neighboring blocks or recently coded level magnitudes). As described in Section VI, we are currently working on the ratedistortion optimization of the graph that would allow to select the best coding option among all these tools.…”

Visual data compression for multimedia applications

“…This process will reduce the number of code words needed to code nonzero coefficients. For high-compression subband coding [35] (compression ratios higher than 50), it is of major importance to exploit the existing zero correlation across the subbands as proposed in [36]- [38] in order to maintain a good quality. Fig.…”

<title>Evaluation of color-embedded wavelet image compression techniques</title>

“…Zerotrees take advantage of structural similarities across bands when using wavelet decomposition of an image. A shortcoming of Shapiro's [2,10] algorithm is that it addresses the compression of monochrome images only.…”

“…The use of spatial orientation trees (SOTs), such as zerotrees, introduced first by Shapiro [2,10], resulted in a very efficient method for achieving rate scalability. A variation of this algorithm, referred to as Set Partitioning in Hierarchical Trees (SPIHT), was proposed by Said and Pearlman [1].…”