1 of 1

VDL-Surrogate: A View-Dependent Latent-based Model for Parameter Space Exploration of Ensemble Simulations (IEEE VIS’22 Best Paper Honorable Mention)

Scientific Achievement

Enable scientists to explore the parameter space of simulations without running the simulation from all possible simulation parameters.

Significance and Impact

A view-dependent latent representation approach to support parameter space exploration with high-resolution visualization results and user-specified visual mappings.

Comparison of the images generated using VDL-Surrogate and InSituNet for the Nyx dataset with the ground truth images for the MPAS-Ocean (left) and Nyx (right) dataset.

Technical Approach

An autoencoder called Ray AutoEncoder is trained to encode samples along each ray with a latent representation
For a selected viewpoint, we train a model called View-Dependent Latent Predictor to learn the mapping between simulation parameters and view-dependent latent representaions
Scientists can predict the view-dependent latent representations, decode the latent representations, and perform visualization using user-specified visual mappings

Neng Shi, Jiayi Xu, Haoyu Li, Hanqi Guo, Jonathan Woodring, and Han-Wei Shen: VDL-Surrogate: A View-Dependent Latent-based Model for Parameter Space Exploration of Ensemble Simulations, IEEE Transactions on Visualization and Computer Graphics (Proc. IEEE VIS 2022), 229(1), 820-830, 2023. [Best Paper Honorable Mention Award at IEEE VIS 2022]

In this work, we propose a view-dependent neural-network-latent-based surrogate model for parameter space exploration of ensemble simulations that allows high-resolution visualizations and user-specified visual mappings. Surrogate-enabled parameter space exploration allows domain scientists to preview simulation results without having to run a large number of computationally costly simulations. Limited by computational resources, however, existing surrogate models may not produce previews with sufficient resolution for visualization and analysis. To improve the efficient use of computational resources and support high-resolution exploration, we perform ray casting from different viewpoints to collect samples and produce compact latent representations. This latent encoding process reduces the cost of surrogate model training while maintaining the output quality. In the model training stage, we select viewpoints to cover the whole viewing sphere and train corresponding models for the selected viewpoints. In the model inference stage, we predict the latent representations at previously selected viewpoints and decode the latent representations to data space. For any given viewpoint, we make interpolations over decoded data at selected viewpoints and generate visualizations with user-specified visual mappings. Quantitative and qualitative evaluations in cosmological and ocean simulations shows the effectiveness and efficiency of our approach.