IEEE VIS 2024 Content: A Comparative Study of Neural Surface Reconstruction for Scientific Visualization

A Comparative Study of Neural Surface Reconstruction for Scientific Visualization

Siyuan Yao - University of Notre Dame, Notre Dame, United States

Weixi Song - Wuhan University, Wuhan, China

Chaoli Wang - University of Notre Dame, Notre Dame, United States

Screen-reader Accessible PDF
Download preprint PDF
Download camera-ready PDF
Download Supplemental Material

Room: Bayshore VI

2024-10-16T16:27:00ZGMT-0600Change your timezone on the schedule page
2024-10-16T16:27:00Z
Exemplar figure, described by caption below
We selected 10 representative surface reconstruction methods and created 9 datasets for evaluation. Each dataset comprises 42 images for training and 181 images for testing. After training the models, we used them to generate neural surface rendering images and reconstruct surface polygon meshes. The synthesized results were evaluated using peak signal-to-noise ratio (PSNR), learned perceptual image patch similarity (LPIPS) against ground truth images, and chamfer distance against the ground truth surface mesh. We also comprehensively analyzed the results, including model design and performance.
Fast forward
Keywords

Machine Learning Techniques, Datasets

Abstract

This comparative study evaluates various neural surface reconstruction methods, particularly focusing on their implications for scientific visualization through reconstructing 3D surfaces via multi-view rendering images. We categorize ten methods into neural radiance fields and neural implicit surfaces, uncovering the benefits of leveraging distance functions (i.e., SDFs and UDFs) to enhance the accuracy and smoothness of the reconstructed surfaces. Our findings highlight the efficiency and quality of NeuS2 for reconstructing closed surfaces and identify NeUDF as a promising candidate for reconstructing open surfaces despite some limitations. By sharing our benchmark dataset, we invite researchers to test the performance of their methods, contributing to the advancement of surface reconstruction solutions for scientific visualization.