Finding a meaningful correspondence between two or more shapes is one of the most fundamental shape analysis tasks. The problem can be generally stated as: given input shapes S1,S2,...,SN, find a meaningful relation (or mapping) between their elements. Under different contexts, the problem has also been referred to as registration, alignment, or simply, matching. Shape correspondence is a key algorithmic component in tasks such as 3D scan alignment and space-time reconstruction, as well as an indispensable prerequisite in diverse applications including attribute transfer, shape interpolation, and statistical modeling.
데이터셋
출처에 같은 태스크 이름으로 연결된 데이터셋입니다. 동명 태스크가 포함될 수 있으므로, 비교할 때는 평가 결과의 조건도 확인하세요.
Shape matching plays an important role in geometry processing and shape analysis. In the last decades, much research has been devoted to improve the quality of matching between surfaces. This huge effort is motivated by several applications such as object retrieval, animation and information transfer just to name a few. Shape matching is usually divided into two main categories: rigid and non rigid matching. In both cases, the standard evaluation is usually performed on shapes that share the same connectivity, in other words, shapes represented by the same mesh. This is mainly due to the availability of a “natural” ground truth that is given for these shapes. Indeed, in most cases the consistent connectivity directly induces a ground truth correspondence between vertices. However, this standard practice obviously does not allow to estimate the robustness of a method with respect to different connectivity. With this track, we propose a benchmark to evaluate the performance of point-to-point matching pipelines when the shapes to be matched have different connectivity (see Figure 1). We consider the concurrent presence of 1) different meshing, 2) rigid transformation in 3D space, 3) non-rigid deformations, 4) different vertex density, ranging from 5K to more than 50K, and 5) topological changes induced by mesh gluing in areas of contact. The correspondence between these shapes is obtained through the recently proposed registration pipeline FARM [1]. This method provides a high-quality registration of the SMPL model [2] to a large set of human meshes coming from different datasets from which we obtain a well-defined correspondence for all the meshes registered and SMPL itself.
Hi-resolution three-dimensional nonrigid shapes in a variety of poses for non-rigid shape similarity and correspondence experiments. The database contains a total of 80 objects, including 11 cats, 9 dogs, 3 wolves, 8 horses, 6 centaurs, 4 gorillas, 12 female figures, and two different male figures, containing 7 and 20 poses. Typical vertex count is about 50,000. Objects within the same class have the same triangulation and an equal number of vertices numbered in a compatible way. This can be used as a per-vertex ground truth correspondence in correspondence experiments. Two representations are available: MATLAB file (.mat) and ASCII text files containing the 1-based list of triangular faces (.tri), and a list of vertex XYZ coordinates (.vert). A .png thumbnail is available for each object.