Introducing Stable Point Aware 3D: Real-Time Editing and Complete Object Structure Generation
Key Takeaways:
Stable Point Aware 3D (SPAR3D) introduces real-time editing and complete structure generation of a 3D object from a single image in less than a second.
With a first-of-its-kind architecture, SPAR3D combines precise point cloud sampling with advanced mesh generation to deliver unprecedented control over 3D asset creation. To learn more about the underlying technology, you can read the full research paper here.
This model is free for both commercial and non-commercial use under the permissive Stability AI Community License.
You can download the weights on Hugging Face and code on GitHub, or access the model through the Stability AI Developer Platform API.
As announced in partnership with NVIDIA at CES, we’re excited to introduce Stable Point Aware 3D (SPAR3D), setting a new standard in 3D generation. Ideal for running on NVIDIA RTX AI PCs, SPAR3D enables real-time editing and complete structure generation of 3D objects from a single image in under a second.
You can download the weights on Hugging Face and code on GitHub, or access the model through the Stability AI Developer Platform API.
Where the model excels
SPAR3D transforms 3D prototyping for game developers, product designers, and environment builders with the following advanced capabilities:
Unprecedented Control: Allows users to directly edit the point cloud by deleting, duplicating, stretching, adding features, or recoloring points.
Complete Structure Prediction: Enhances 3D construction by providing accurate geometry and detailed predictions for full 360-degree views, including areas typically hidden, such as the back of an object.
Lightning-Fast Generation: Converts edited point clouds into final meshes in just 0.3 seconds, enabling seamless real-time editing. From a single input image, it generates highly detailed 3D meshes in only 0.7 seconds per object.
A first-of-its-kind architecture
SPAR3D's architecture combines precise point cloud sampling with advanced mesh generation to deliver unprecedented control over 3D asset creation. The process works in two stages:
First Stage: A specialized point diffusion model generates a detailed point cloud capturing the object's fundamental structure.
Second Stage: The triplane transformer processes this point cloud alongside the original image features, producing high-resolution triplane data. This data drives the final reconstruction, accurately capturing geometry, texture, and illumination from the source image.
This dual-stage approach uniquely combines regression-based modeling's precision with generative techniques' flexibility, enabling both accurate reconstruction and creative control. This advanced architecture ensures precise geometry and accurate 360-degree predictions, streamlining 3D modeling workflows.
To learn more about the underlying technology, you can read the full research paper here.
SPAR3D Two-Stage Architecture Diagram
Get started
SPAR3D is now free for both commercial and non-commercial use under the permissive Stability AI Community License. For organizations with annual revenue more than $1M, please contact us here to inquire about an Enterprise License.
You can download SPAR3D from Hugging Face, access the code on GitHub, and integrate it through the Stability AI Developer Platform API.
To stay updated on our progress follow us on X, LinkedIn, Instagram, and join our Discord Community.
