3D Object Categorization Using Spin-Images with MPI Parallel Implementation

Document Type : Original Article


Computer Systems Department, Ain Shams University


Object recognition and categorization are two important key features of computer vision. Accuracy aspects
representresearchchallengeforbothobjectrecognitionandcategorization techniques. High performance computing
(HPC) technologies usually manage the increasing time and complexity of computations. In this paper, an approach
utilizing3Dspin-imagesfor3Dobjectcategorizationisproposed.Themaincontributionofthisapproach is that item ploys the MPI
techniques in a unique way to extract spin-images. The technique proposed utilizes the independence between spin-images generated a teach point. Time estimation of this technique has shown dramatic decrease of the categorization time proportion alto number of workers used.