To facilitate processing of 3D objects is common to use high-level representations. The interest points are one of them. An interest point should possess a distinctive feature regarding its locality and should be stable in different instances of the object. This article proposes a descriptor based on symmetry (GISIFs) and heat diffusion (HKS). From this features, we select a set of representative points. The GISIFs referenced in this article has not been used to extract local features. We compare our results with the results of other techniques, which make up the state of the art in interest point detection. We use a benchmark that evaluates the accuracy of the selected points with respect to an ideal set of interest points.

In order to facilitate 3D object processing, it is common to use high-level representations such as local descriptors that are usually computed using defined neighborhoods. K-rings, a technique to define them, is widely used by several methods. In this work, we propose a model for the distributed computation of local descriptors over 3D triangular meshes, using the concept of k-rings. In our experiments, we measure the performance of our model on huge meshes, evaluating the speedup, the scalability, and the descriptor computation time. We show the optimal configuration of our model for the cluster we implemented and the linear growth of computation time regarding the mesh size and the number of rings. We used the Harris response, which describes the saliency of the object, for our tests.

In order to facilitate 3D object processing, it is common to use high-level representations such as local descriptors that are usually computed using defined neighborhoods. K-rings, a technique to define them, is widely used by several methods. In this work, we propose a model for the distributed computation of local descriptors over 3D triangular meshes, using the concept of k-rings. In our experiments, we measure the performance of our model on huge meshes, evaluating the speedup, the scalability, and the descriptor computation time. We show the optimal configuration of our model for the cluster we implemented and the linear growth of computation time regarding the mesh size and the number of rings. We used the Harris response, which describes the saliency of the object, for our tests.