Microsoft has quietly introduced an innovative artificial intelligence feature for its Copilot tool, enabling users to transform static 2D images into 3D models. Named Copilot 3D, this experimental functionality is now accessible through Copilot Labs. The company announced that this feature will be available to all global users, provided they are logged into their personal Microsoft accounts. While Microsoft is not the pioneer in this area, it maintains that the 3D models generated with Copilot are ready for printing and rendering.
Copilot 3D Is Now Available to All Users
Although there was no formal announcement concerning this new feature, Microsoft has released a detailed blog post. The company characterizes Copilot 3D as an AI-driven tool that enables users to convert a 2D image into a meticulously rendered 3D model. As an experimental function available through Copilot Labs, it does not require any paid subscription.
According to Microsoft, this AI tool simplifies the 3D asset creation process, making it accessible for individuals without prior knowledge of 3D modeling. Users can upload images up to 10MB in size, and Copilot 3D will process these into models that are suitable for both printing and rendering. The generated outputs can be directly utilized in gaming, animation, 3D printing, augmented reality (AR), virtual reality (VR), and various metaverse-oriented digital content projects.
For those unsure of where to begin, Microsoft provides a curated library of unique scenes to inspire users. Importantly, Copilot 3D currently focuses exclusively on converting 2D images into 3D models, with no capability for text-to-3D transformations at this time. The company suggests users access this feature via the desktop version of Copilot, while it is also compatible with mobile browsers.
While Microsoft has made strides in providing an easy-to-use approach for creating 3D models from 2D images, it is not the first in the market. Last year, Stability AI launched an open-source AI model known as Stable Video 3D (SV3D) that offered similar functionality. Additionally, tech giants Tencent and Meta unveiled their own 3D asset generation models earlier this year. Meanwhile, researchers from Apple introduced an open-source model called Matrix3D, capable of converting images into 3D forms.
