MAKE-IT-FAB: Modeling of Shapes for Personal Fabrication

The aim of this project is to investigate and to contribute to shape modeling and geometry processing for personal fabrication---a trend that currently receives intensified attention in science and industry. This wide and complex topic inherently spans multiple scientific fields, but one of the major areas of research are intelligent algorithms for shape processing. We argue that if models need to be printed, this goal should be considered from the beginning of the model design process. This research falls into the area of computer graphics and interactive techniques where we want to build on top of the insights known from interactive and procedural modeling, shape analysis and understanding, as well as geometry processing. Our goal is to contribute novel algorithmic solutions to the fabrication-aware shape processing and interactive modeling. In particular, we want to research following aspects: (1) One of the major problems of modern shape modeling approaches is the question of high-level shape understanding, i.e., how complex composite shapes should be interpreted, organized, hierarchized, divided, or grouped into their particular parts. (2) Further, we want to research novel interactive shape modeling methods for creative making of esthetically appealing shapes. We will develop a high-level modeling application which aims co-called exploratory shape modeling. It should be well suited for occasional and non-professional users and help them to set their creativity free, while supporting them during the modeling process. (3) Finally, we will provide a set of optimization and simulation techniques in order to integrate the requirements given by consumer-level desktop fabrication directly into a fabrication-aware modeling process. This tight integration allows to immediately create 3D printable models without any intermediate stages, as opposed to the current work flow. The research results of this project will impact a large audience in both science and industry. We expect to directly contribute to the scientific state-of-the-art in computer graphics and interactive techniques. In particular, we will contribute to the shape understanding problem, the shape synthesis field, the interactive modeling field, and finally to the shape optimization and physical modeling field. Additionally, our results will have an indirect impact in education, consumer-level design, entertainment industries, and eventually it might also impact professional design areas, like product design, rapid prototyping, or architectural design. From the economical point of view, our research will impact the consumer additive fabrication market, aka 3D printing, which is currently rapidly growing.

The aim of the project was to study and improve solutions for the problem of creating digital models for personal manufacturing such as 3D-printing. The paradigm of making custom objects has recently received increasing attention in science and industry. This broad and complex topic encompasses several scientific areas, but one of the research topics is algorithms for shape optimization, which were also the focus of our research. Over the five years of the project, we have published a total of 9 articles in international journals. It should be noted that 3 of the articles were featured at the ACM SIGGRAPH conference and published in ACM Transactions on Graphics, the highest rated journal in our scientific field. In addition, we have completed 6 academic theses: 1 doctoral thesis, 3 master theses and 2 bachelor theses. A second doctoral thesis is about to be completed. Eventually the Principal Investigator was promoted to Associate Professor at the New Jersey Institute of Technology in the United States. Our proposed solutions for shape optimization were the first to introduce shape optimization for 3D printing with smooth offset surfaces. Since 2015, the publications have been cited more than 100 times according to Google Scholar, and the proposed methods have laid the foundation for several subsequent work by the project team itself and other researchers. We have introduced methods for optimizing shapes for 3D-printing, CNC-milling, and for manufacturing with laser-cutters and industrial robots. The proposed algorithms are aimed at computational efficiency and ease of use so that they can be used by inexperienced users as well as professional designers. They include solutions for computer-aided design of individual comfortable seats, or methods for the automatic calculation of so-called artistic "string art" images, which consist of kilometers of threads. During the project we worked with international colleagues from other universities as well as with internal colleagues and students. We also worked on an interdisciplinary basis with architecture and design students at the TU Wien. In addition to the scientific field, the results of the project met with great media interest. In 2015, the project won first prize at the Austrian Computer Graphics Award in the "Best Technical Solution" category, and one of the publications was awarded the "Best Paper Award Honorable Mention". It has also found an echo in the national and international press, including "Die Presse" and "Der Standard", the two leading Austrian daily newspapers. In addition, many specialized online magazines and forums reported on the results of the project. Overall, the goal of researching calculation methods for personal production was achieved and the results of the project provide a solid basis for further research as well as for further practical applications.