Out‐of‐core compression and decompression of large n‐dimensional scalar fields

Abstract

We present a simple method for compressing very large and regularly sampled scalar fields. Our method is particularlyattractive when the entire data set does not fit in memory and when the sampling rate is high relative to thefeature size of the scalar field in all dimensions. Although we report results for and data sets, the proposedapproach may be applied to higher dimensions. The method is based on the new Lorenzo predictor, introducedhere, which estimates the value of the scalar field at each sample from the values at processed neighbors. The predictedvalues are exact when the n‐dimensional scalar field is an implicit polynomial of degree n− 1. Surprisingly,when the residuals (differences between the actual and predicted values) are encoded using arithmetic coding,the proposed method often outperforms wavelet compression in anL∞sense. The proposed approach may beused both for lossy and lossless compression and is well suited for out‐of‐core compression and decompression,because a trivial implementation, which sweeps through the data set reading it once, requires maintaining only asmall buffer in core memory, whose size barely exceeds a single (n−1)‐dimensional slice of the data. Categories and Subject Descriptors (according to ACM CCS): I.3.5 [Computer Graphics]: Compression, scalar fields,out‐of‐core.

Lossless Compression of Hexahedral Meshes

“…To make the best possible use of available neighbors, we employ spectral prediction [16] by precomputing "optimal" rational weights for all 2 7 = 128 possible configurations (Figure 3). Our derivation shows parallelogram [4] and Lorenzo prediction [6,17] to be special cases (0x70 respectively 0x7f in Figure 3) of spectral prediction.…”

Assessing the effects of data compression in simulations using physically motivated metrics

“…Both compressors rely on polynomial interpolation for prediction, with fixed integer polynomial coefficients. APAX uses univariate Lagrange polynomials of degree 0 and 1, allowing linear polynomials (or any function with fpzip exploits correlations in more than one dimension using the Lorenzo predictor [17], which over a 3D domain reproduces trivariate quadratic polynomials (or any function for which…”

Lossless Compression of Double-Precision Floating-Point Data for Numerical Simulations: Highly Parallelizable Algorithms for GPU Computing

“…al. presented an arithmetic-coding-based method for compressing 3D geometric data [5]- [7]. In their method, we first predict coordinates of a point in a 3D mesh model, each of which is expressed in FP numbers, from coordinates of the adjacent points by some techniques, typically by parallelogram predictors.…”

“…al. presented an arithmetic-coding-based method for compressing 3D geometric data [5]- [7]. Harada et.…”