.png){kind=icon}
.png){kind=icon}
Return
Information Theory-Based Automatic Multimodal Transfer Function Design
.png){kind=icon}
OA
AI
DOI：10.1109/JBHI.2013.2263227.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this paper, we present a new framework for multimodal volume visualization that combines several information-theoretic strategies to define both colors and opacities of the multimodal transfer function. To the best of our knowledge, this is the first fully automatic scheme to visualize multimodal data. To define the fused color, we set an information channel between two registered input datasets, and afterward, we compute the informativeness associated with the respective intensity bins. This informativeness is used to weight the color contribution from both initial 1-D transfer functions. To obtain the opacity, we apply an optimization process that minimizes the informational divergence between the visibility distribution captured by a set of viewpoints and a target distribution proposed by the user. This distribution is defined either from the dataset features, from manually set importances, or from both. Other problems related to the multimodal visualization, such as the computation of the fused gradient and the histogram binning, have also been solved using new information-theoretic strategies. The quality and performance of our approach are evaluated on different datasets.
Keywords:
Information theory
Kullback-Leibler distance
multimodal fusion
multimodal visualization
transfer function design
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.8
Papers:
4.5K
Citations:
2.0W