A General Framework for Augmenting Lossy Compressors with Topological Guarantees

TL;DR

This paper presents a versatile framework that enhances existing lossy data compressors to ensure the preservation of topological features like contour trees, crucial for scientific analysis, by quantifying and encoding necessary data adjustments.

Contribution

It introduces a general method to augment any lossy compressor with topological guarantees, applicable to both traditional and deep learning-based compressors.

Findings

Successfully augmented classic compressors with topological guarantees.

Demonstrated preservation of contour trees in compressed scientific data.

Applicable to various compressor types, including neural network-based methods.

Abstract

Topological descriptors such as contour trees are widely utilized in scientific data analysis and visualization, with applications from materials science to climate simulations. It is desirable to preserve topological descriptors when data compression is part of the scientific workflow for these applications. However, classic error-bounded lossy compressors for volumetric data do not guarantee the preservation of topological descriptors, despite imposing strict pointwise error bounds. In this work, we introduce a general framework for augmenting any lossy compressor to preserve the topology of the data during compression. Specifically, our framework quantifies the adjustments (to the decompressed data) needed to preserve the contour tree and then employs a custom variable-precision encoding scheme to store these adjustments. We demonstrate the utility of our framework in augmenting classic compressors (such as SZ3, TTHRESH, and ZFP) and deep learning-based compressors (such as Neurcomp) with topological guarantees.