arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.01831cs.CVcs.CGcs.LG
收藏

POCO:点卷积曲面重构

POCO: Point Convolution for Surface Reconstruction

LIGM(巴黎东部马恩拉瓦雷计算机与图像实验室);Ecole des Ponts(巴黎桥路学院);Univ Gustave Eiffel(古斯塔夫·埃菲尔大学);CNRS(法国国家科学研究中心)

Alexandre Boulch, Renaud Marlet

摘要:

隐式神经网络已成功地用于点云数据的曲面重建。然而,他们中的许多人面临的可扩展性问题,因为他们编码的等值面函数的整个对象或场景到一个单一的潜在向量。为了克服这一限制,一些方法在粗糙的规则3D网格或3D补丁上推断潜在向量,并对它们进行插值以回答占用查询。在这样做的过程中,它们失去了与物体表面上采样的输入点的直接联系,并且它们在空间中均匀地附加信息,而不是在最重要的地方,即,靠近地表此外,依赖于固定的补丁大小可能需要离散化调整。为了解决这些问题,我们建议使用点云卷积并计算每个输入点的潜在向量。然后,我们使用推断的权重对最近邻执行基于学习的插值。在对象和场景数据集上的实验表明,我们的方法在大多数经典指标上显着优于其他方法,产生更精细的细节和更好地重建更薄的体积。该代码可在https://github.com/valeoai/POCO上获得。

英文摘要:

Implicit neural networks have been successfully used for surface reconstruction from point clouds. However, many of them face scalability issues as they encode the isosurface function of a whole object or scene into a single latent vector. To overcome this limitation, a few approaches infer latent vectors on a coarse regular 3D grid or on 3D patches, and interpolate them to answer occupancy queries. In doing so, they loose the direct connection with the input points sampled on the surface of objects, and they attach information uniformly in space rather than where it matters the most, i.e., near the surface. Besides, relying on fixed patch sizes may require discretization tuning. To address these issues, we propose to use point cloud convolutions and compute latent vectors at each input point. We then perform a learning-based interpolation on nearest neighbors using inferred weights. Experiments on both object and scene datasets show that our approach significantly outperforms other methods on most classical metrics, producing finer details and better reconstructing thinner volumes. The code is available at https://github.com/valeoai/POCO.

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。

↑