Automatically modeling piecewise planar furniture shapes from unorganized point cloud. (August 2020)

• Record Type: Journal Article
• Authors: Zhao, Junhao; Zong, Chen; Cao, Luming; Chen, Shuangmin; Liu, Guozhu; Xu, Jian; Xin, Shiqing
• Is Part Of: Computers & graphics. Volume 90(2020)
• Page Start: 116
• Page End: 125
• Publication Date: 2020-08
• Subjects: Piecewise planar model -- Clustering -- Regularization -- Shape modeling; Computer graphics -- Periodicals; 006.6
• DOI: 10.1016/j.cag.2020.05.019
• Languages: English
• ISSNs: 0097-8493

Abstract

Highlights: We propose a novel algorithmic pipeline for modeling piecewise planar shapes based on clustering. Our algorithm is computationally efficient and resistant to noise. We give a set of strategies on extracting the polygonal representation of a facet and building the face adjacency graph, which leads to the final compact planarization representation. We also introduce a regularization technique to encourage the orthogonality between facing directions, which is further transformed into a purely convex optimization problem. Graphical abstract: Abstract: Piecewise planar 3D objects are very common in digital furniture manufacturing. In this paper, we propose a novel method for automatic reconstruction of 3D objects with planar facets from unorganized point clouds. We formulate this problem into a point clustering problem where the key difficulty lies in consolidating co-planar points into a cluster. In order to achieve this purpose, the first step is to triangulate the input point cloud into a mesh (may have over connectivity) that is a super-set of the underlying manifold mesh surface. Then the hundreds of thousands of normal vectors of triangles, after being mapped onto a Gauss sphere, are capable of reporting reliable facing directions of the faces of the final 3D model. After grouping the points based on co-planarity, we fit each cluster with a planar facet and then assemble them into a piecewise planarization representation of the whole model. We further introduce an additional regularization term to meet the orthogonality requirement on a particular occasion, and then transform this problem into a purely convex optimization problem. Our method is efficient and requires just a few parameters. Extensive experimental results show that it is able to handle point clouds with various levels of noise and yield a desirable piecewise planar 3D model with a clean and compact representation.