Interactive multiscale tensor reconstruction for multiresolution volume visualization

Susanne K. Suter*, Jose A.Iglesias Guitian, Fabio Marton, Marco Agus, Andreas Elsener, Christoph P.E. Zollikofer, M. Gopi, Enrico Gobbetti, Renato Pajarola

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

44 Scopus citations


Large scale and structurally complex volume datasets from high-resolution 3D imaging devices or computational simulations pose a number of technical challenges for interactive visual analysis. In this paper, we present the first integration of a multiscale volume representation based on tensor approximation within a GPU-accelerated out-of-core multiresolution rendering framework. Specific contributions include (a) a hierarchical brick-tensor decomposition approach for pre-processing large volume data, (b) a GPU accelerated tensor reconstruction implementation exploiting CUDA capabilities, and (c) an effective tensor-specific quantization strategy for reducing data transfer bandwidth and out-of-core memory footprint. Our multiscale representation allows for the extraction, analysis and display of structural features at variable spatial scales, while adaptive level-of-detail rendering methods make it possible to interactively explore large datasets within a constrained memory footprint. The quality and performance of our prototype system is evaluated on large structurally complex datasets, including gigabyte-sized micro-tomographic volumes.

Original languageEnglish (US)
Article number6064978
Pages (from-to)2135-2143
Number of pages9
JournalIEEE Transactions on Visualization and Computer Graphics
Issue number12
StatePublished - Nov 15 2011


  • interactive volume visualization
  • multiresolution rendering
  • multiscale
  • tensor reconstruction

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Computer Graphics and Computer-Aided Design


Dive into the research topics of 'Interactive multiscale tensor reconstruction for multiresolution volume visualization'. Together they form a unique fingerprint.

Cite this