arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2402.14009cs.LGcs.CV
收藏

几何信息神经网络

Geometry-Informed Neural Networks

LIT AI Lab, Institute for Machine Learning, JKU Linz, Austria(LIT人工智能实验室,机器学习研究所,JKU林茨,奥地利);Emmi AI GmbH, Linz, Austria(Emmi人工智能公司,林茨,奥地利);Department of Mathematics, University of Oslo, Norway(数学系,奥斯陆大学,挪威)

Arturs Berzins, Andreas Radler, Eric Volkmann, Sebastian Sanokowski, Sepp Hochreiter, Johannes Brandstetter

AI总结:

本文提出几何信息神经网络(GINNs),通过将用户设计目标和多样性约束引入无数据训练框架,实现形状生成并避免模式坍塌,在物理、几何和工程设计任务中验证了对几何与拓扑性质的控制能力。

摘要:

几何是计算机图形学、设计和工程中普遍使用的工具。然而,大型形状数据集的缺乏限制了最先进监督学习方法的应用,并促使人们探索替代学习策略。为此,我们引入了几何信息神经网络(GINNs)——一个无需数据即可训练形状生成神经场的框架,通过利用用户指定的设计需求(以目标和约束的形式)。通过将多样性作为显式约束,GINNs避免了模式坍塌,并能生成多个多样化的解,这在几何任务中通常是必需的。在实验中,我们将GINNs应用于涵盖物理、几何和工程设计的多个问题,展示了对几何和拓扑性质的控制,例如表面光滑度或孔的数量。这些结果证明了无需数据训练形状生成模型的潜力,为无需大型数据集的新生成式设计方法铺平了道路。

英文摘要:

Geometry is a ubiquitous tool in computer graphics, design, and engineering. However, the lack of large shape datasets limits the application of state-of-the-art supervised learning methods and motivates the exploration of alternative learning strategies. To this end, we introduce geometry-informed neural networks (GINNs) -- a framework for training shape-generative neural fields without data by leveraging user-specified design requirements in the form of objectives and constraints. By adding diversity as an explicit constraint, GINNs avoid mode-collapse and can generate multiple diverse solutions, often required in geometry tasks. Experimentally, we apply GINNs to several problems spanning physics, geometry, and engineering design, showing control over geometrical and topological properties, such as surface smoothness or the number of holes. These results demonstrate the potential of training shape-generative models without data, paving the way for new generative design approaches without large datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑