arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.15529cs.LGcs.CVstat.ML
收藏

TT-NF:张量训练神经场

TT-NF: Tensor Train Neural Fields

ETH Zürich(苏黎世联邦理工学院);KU Leuven(鲁汶大学)

Anton Obukhov, Mikhail Usvyatsov, Christos Sakaridis, Konrad Schindler, Luc Van Gool

摘要:

学习神经场一直是深度学习研究中的一个活跃话题,除其他问题外,重点是寻找更紧凑且易于拟合的表示。在本文中,我们介绍了一种新的低秩表示称为张量训练神经场(TT-NF)的密集规则网格上学习神经场和有效的方法,从他们的采样。我们的表示是神经场的TT参数化,用反向传播训练以最小化非凸目标。我们分析了低秩压缩对下游任务质量指标的影响,在两种设置。首先,我们证明了我们的方法的效率在沙箱任务的张量去噪,承认比较基于SVD的计划,旨在最大限度地减少重建误差。此外,我们将所提出的方法应用于神经辐射场,其中对应于最佳质量的场的低秩结构只能通过学习来发现。

英文摘要:

Learning neural fields has been an active topic in deep learning research, focusing, among other issues, on finding more compact and easy-to-fit representations. In this paper, we introduce a novel low-rank representation termed Tensor Train Neural Fields (TT-NF) for learning neural fields on dense regular grids and efficient methods for sampling from them. Our representation is a TT parameterization of the neural field, trained with backpropagation to minimize a non-convex objective. We analyze the effect of low-rank compression on the downstream task quality metrics in two settings. First, we demonstrate the efficiency of our method in a sandbox task of tensor denoising, which admits comparison with SVD-based schemes designed to minimize reconstruction error. Furthermore, we apply the proposed approach to Neural Radiance Fields, where the low-rank structure of the field corresponding to the best quality can be discovered only through learning.

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。

↑