Computing Digital Signature by Transforming 2D Image to 3D: A Geometric Perspective

Abstract

Recently 2D to 3D face reconstruction has attracted a lot of interest in the area of Computer vision. There has been a recent increase in research into various 3D reconstruction techniques using neural nets, with the majority of approaches producing high-quality results and efficiency. This paper presents an approach to convert 2D facial images to 3D and then use the 3D data and features to construct a unique digital signature. The proposed solution eliminates the need for several pictures and reduces the calculation load. The main objective of this research is to understand the progress that has already been made in this domain and we came up with an open question of whether the generation of a face mesh is possible using a Feature Vector. This would reduce the storage space required for the 3D Facial data. If Face mesh reconstruction is possible using the feature vector only, it would drastically reduce time and space for various 2D and 3D image analysis applications. The feature vector would later be compressed to create a unique digital signature. Accessing any information on the database with a key will be efficient and fast.