I. INTRODUCTION

II. PREVIOUS WORK

III. OVERVIEW

IV. 2TO3SSC DESCRIPTOR GENERATION

V. COMPONENT MATCHING

VI. RESULTS AND DISCUSSION

VII. CONCLUSION

REFERENCES

We present a method to compute the descriptor of components of point clouds, therefore, a novel component-oriented partial matching of point clouds is achieved based on the component descriptor. We observe that 3D components can be constructed by stacking 2D shapes using certain criteria so that the centers of the 2D shapes form a curve called a skeletal curve that is the trajectory of the 2D shapes. In addition, the scaling factors of the 2D shapes also impact the shape of the 3D components. Motivated by these observations, the computation of the component descriptor that is termed 2to3SSC (from 2D to 3D: 2D Shape and Skeletal Curve) is formulated as a 2D shape and skeletal curve extraction problem, and the component-oriented partial matching of the point clouds is based on the dissimilarity measure of 2to3SSCs of the components. Furthermore, for the 2D shape matching, which is crucial to the matching of the components, we present a novel 2D shape descriptor called VDTL (Vertical Distances to the Tangent Line). The proposed method outperforms previously proposed methods because it simultaneously encodes the local and global features of the components as opposed to only encoding the local or partial features as in previous studies. Finally, the effectiveness and performance of 2to3SSCs are compared with those of stateof-the-art feature description and matching methods for different point cloud datasets. Further, the benefits and the applicability of the proposed method are demonstrated; favorable results are obtained for real-world point clouds of Terracotta fragments.

INTRODUCTION

Feature extraction, description, and matching are the core and prerequisite of most point cloud processing techniques, such as point cloud registration [1], line drawings generated from point clouds [2], 3D object retrieval [3], 3D object partitioning [4], and 3D object reconstruction [5]. It is therefore not surprising that numerous studies have reported on techniques for addressing the problem of feature extraction, description, and matching of point clouds [3], [6]. However, the main differences among these techniques are the geometric scale at which the analysis is performed and the matching rules (or similarity measures) for the corresponding feature descriptors. Thus, the features referenced in previous studies can be roughly classified into three categories related to the geometric scale: micro-features, meso-features, and macro-features, as shown in Fig. 1. Geometric operators at a fine scale (referred to as micro-features in this paper) concentrate on extracting small structures and capturing details that describe the gradients of the point clouds, as shown in Fig.1 (b). The extracted features are scattered points or line segments. These methods are often applied to (i) the generation of line drawings in non-photorealistic rendering (NPR) [7] since the line drawings convey the shape of the models; (ii) the reassembly of fragments based on the matching of feature points [8] or feature lines [5]; (iii) as a pre-requisite step for generating the meso-feature descriptors [6].