#### Introducing Style2Fab: AI-Driven Tool for Customizing 3D Models
As 3D printers become more affordable and accessible, more people are using them to create their own objects. However, adding personalized design elements to 3D models can be challenging, especially for novice makers. Traditional computer-aided design (CAD) software is complex and expensive, and it can be difficult to modify models that are not available online. Additionally, ensuring that customizations do not affect the functionality of the object requires domain expertise that many beginners lack.
To address these challenges, researchers at MIT have developed Style2Fab, an AI-driven tool that allows users to add custom design elements to 3D models without compromising their functionality. This tool simplifies the process by enabling users to describe their desired design using natural language prompts. The models can then be fabricated using a 3D printer.
Style2Fab uses deep-learning algorithms to automatically separate the model into aesthetic and functional segments, making it easier for users to customize the design.
#### Making 3D Printing and Personalization Accessible
Style2Fab not only empowers novice designers, but it also has potential applications in medical making. Research has shown that considering both the aesthetic and functional aspects of assistive devices increases the likelihood that patients will use them. However, clinicians and patients may not have the expertise to personalize 3D-printable models. Style2Fab solves this problem by allowing users to customize the appearance of medical devices without compromising their functionality.
The researchers behind Style2Fab, including lead author Faraz Faruqi and co-senior author Stefanie Mueller, believe that providing a user-friendly tool for DIY assistive technology is crucial.
#### The Functionality of Style2Fab
To develop Style2Fab, the researchers studied existing 3D models available in online repositories like Thingiverse. They aimed to understand the different functionalities within these models in order to effectively segment them. This analysis revealed two main functionalities: external functionality, which relates to parts of the model that interact with the outside world, and internal functionality, which involves parts that need to fit together after fabrication.
Style2Fab uses machine learning to analyze the model’s topology and identify changes in geometry. Based on this analysis, it divides the model into segments. These segments are then compared to a dataset of 3D objects, allowing Style2Fab to classify them as functional or aesthetic.
#### Styling and Personalizing 3D Models
Once the segmentation process is complete, users can describe their desired design elements using natural language prompts. An AI system called Text2Mesh then generates a modified 3D model that meets the user’s criteria. This system manipulates the aesthetic segments of the model and adds texture, color, or adjusts shape to achieve the desired result. However, the functional segments remain untouched to maintain the object’s intended functionality.
The user interface of Style2Fab simplifies the process, allowing users to segment and stylize the 3D model with a few clicks and inputs.
#### User Study and Future Developments
The researchers conducted a study with makers of varying experience levels in 3D modeling. They found that Style2Fab was useful for both novice and experienced users. Novice users were able to understand and use the interface to customize designs, while experienced users found it helpful in speeding up their workflows.
Moving forward, the researchers aim to further enhance Style2Fab by allowing users to control physical properties as well as geometry. They also plan to add features that enable users to generate their own custom 3D models within the system. The researchers are collaborating with Google on a follow-up project.
Overall, Style2Fab is a powerful tool that makes 3D printing more accessible and customizable, opening up new possibilities for makers and users alike.