MIT researchers have developed Style2Fab, an AI-driven tool that simplifies the customization of 3D printable models, making 3D printing more accessible to novice makers. Many amateur 3D printing enthusiasts rely on free, open-source repositories for models, but adding personal design elements can be challenging and often requires complex CAD software.
Style2Fab streamlines this process by allowing users to describe their desired design using natural language prompts. It utilizes deep-learning algorithms to automatically divide 3D models into aesthetic and functional segments. This division ensures that customizations do not compromise the functionality of the printed objects. For example, a designer may want to modify a part with hinged mechanisms by adding textures with AI. Normally, use of AI might add the textures, but lose the hinges. With Style2Fab, it retains the new aesthetic design elements while keeping the functional elements (the hinges) intact.
The tool was created to empower novice designers and has potential applications in the field of medical making, enabling users to customize assistive devices without affecting their functionality. This accessibility is crucial for patients and clinicians who may lack expertise in 3D modeling.
Style2Fab uses machine learning to classify segments as functional or aesthetic, and it involves users in this classification process, providing initial recommendations that users can adjust. Users then describe their design preferences, and an AI system called Text2Mesh modifies the aesthetic segments accordingly. You can see Style2Fab at work in the video below.
The tool was tested with makers of varying experience levels and proved valuable for both novices and experienced users. For novices, it offered an easy way to stylize designs, while experienced users benefited from quicker workflows and fine-grained control options.
In the future, the researchers plan to enhance Style2Fab to allow control over physical properties and enable users to generate custom 3D models from scratch. They are also collaborating with Google on a follow-up project.
Come and let us know your thoughts on our Facebook, Twitter, and LinkedIn pages, and don’t forget to sign up for our weekly additive manufacturing newsletter to get all the latest stories delivered right to your inbox.
