Stability AI, the company famous for the Stable Diffusion AI image model, has revealed an exciting new technology named TripoSR. This special tool can create top-notch 3D models from just one picture in less than a second.
Teaming up with Tripo AI, Stability AI made TripoSR to meet the needs of professionals in entertainment, gaming, industrial design, and architecture. It quickly provides detailed 3D objects that are easy to visualize.
What makes TripoSR stand out is how fast and easy it is to use. When tried out on an Nvidia A100 GPU, it can produce draft-quality 3D results with textures in only 0.5 seconds. This beats other similar tools like OpenLRM.
But the real strength of TripoSR is its flexibility. Even if you don’t have a super powerful GPU, you can still use it because the model is designed to work with low computing resources, as mentioned in the press release.
Taking inspiration from the recent LRM (Large Reconstruction Model For Single Image to 3D), the Stability AI and Tripo AI teams have added many technical improvements to the basic model. These enhancements include optimizing channel numbers, mask supervision, and a more efficient crop rendering strategy, all contributing to the model’s amazing performance.
Source: Indianexpress
