IEEE VIS 2024 Content: Accelerated Depth Computation for Surface Boxplots with Deep Learning

Accelerated Depth Computation for Surface Boxplots with Deep Learning

Mengjiao Han - University of Utah, Salt Lake City, United States

Tushar M. Athawale - Oak Ridge National Laboratory, Oak Ridge, United States

Jixian Li - University of Utah, Salt Lake City, United States

Chris R. Johnson - University of Utah, Salt Lake City, United States

Screen-reader Accessible PDF

Room: Bayshore VI

2024-10-14T12:30:00ZGMT-0600Change your timezone on the schedule page
2024-10-14T12:30:00Z
Exemplar figure, described by caption below
Functional depth is a valuable technique for analyzing uncertainty of 1D data, and surface boxplots extend this concept to image ensembles, aiding in identifying representative and outlier images. However, the high computational cost limits their usability. This paper introduces a deep-learning framework for efficient surface boxplot computation in time-varying ensemble data. Our method accelerates depth prediction, achieving up to 15X speedups on a GPU while maintaining 99% rank preservation accuracy, making it a practical solution for integrating surface boxplots into visualization tools.
Abstract

Functional depth is a well-known technique used to derive descriptive statistics (e.g., median, quartiles, and outliers) for 1D data. Surface boxplots extend this concept to ensembles of images, helping scientists and users identify representative and outlier images. However, the computational time for surface boxplots increases cubically with the number of ensemble members, making it impractical for integration into visualization tools.In this paper, we propose a deep-learning solution for efficient depth prediction and computation of surface boxplots for time-varying ensemble data. Our deep learning framework accurately predicts member depths in a surface boxplot, achieving average speedups of 6X on a CPU and 15X on a GPU for the 2D Red Sea dataset with 50 ensemble members compared to the traditional depth computation algorithm. Our approach achieves at least a 99\% level of rank preservation, with order flipping occurring only at pairs with extremely similar depth values that pose no statistical differences. This local flipping does not significantly impact the overall depth order of the ensemble members.