PFI:AIR - TT: A System for 3D Content-based Data Management

Drexel University, Philadelphia PA

Investigators

Abstract

This PFI:AIR Technology Translation project focuses on translating 3-Dimensional (3D) content management technology to fill the need for systems that are capable of processing 3D content and sustaining expansion of 3D manufacturing. This 3D Content-based data management system is important because of its impact on information support for the 3D printing and manufacturing industry. The 3D manufacturing industry is facing a growing challenge in managing web-scale 3D content and the progress in managing 3D repositories is still in its early stages. As a result, systems that are capable of managing large-scale 3D content play a critical role in fostering a market for 3D content that is more accessible and offers the potential for new job and income opportunities for both content creators and consumers (e.g. by reducing the barrier for a startup to bring a new product to market). The project will result in a proof-of-concept of a web-scale system with the following unique features: object browsing, query processing, and interaction of 3D object composition/decomposition mechanisms. These features provide the advantages of an implementation based on a micro-services model for complex applications, which is a new architecture in this market space. This project addresses several technology gaps in 3D information management as it translates from research discovery toward commercial application. Specifically, it includes computational methods such as similarity measurements based on a many-to-many matching algorithm that work with volumetric representations of 3D objects and produce a direct correspondence between two objects and their parts, which can be used for registration and juxtaposition. It will also transition several developed mechanisms for indexing a part-based representation of 3D objects, which in turn can be used for efficient retrieval of candidates from a database of models. Our proposal for partial matching of 3D parts will utilize a novel morphing of level sets of 3D surfaces that provides a shape similarity metric between the two surfaces. This metric will allow us to perform high accuracy matching of generalized surfaces. In addition, personnel involved in this project, including four upper-level (junior or senior) undergraduates, will receive entrepreneurship and technology translation experience through exposure to real-world software engineering processes, information retrieval, and large-scale data management tools. This experience will provide the skills, knowledge and experience to prepare the students for either academic careers or employment in industry. The project engages 3D Industries (3DI) Ltd. to ensure the Drexel team?s translational research efforts under the project remain on track to solve critical commercial needs and to minimize the time to market. 3DI's participation will augment the Drexel team?s research efforts while also creating the opportunity for the Drexel team to quickly receive and adapt to feedback from real-work users.