Wagyo

• 

  Wagyo transforms the word "tree" into a prompt that generates a 3D model of a tree.

• 

  Wagyo allows you to transform your spoken or written words into complex 3D designs.

Inspiration

Lego! Wagyo envisions a future where virtual reality becomes a universal canvas. By harnessing the power of AI, we can break down barriers and give everyone the opportunity to express themselves through familiar languages that control new mediums.

What it does

Wagyo aims to revolutionize the way people build virtual reality. With its AI-powered interface, users can create shapes, textures, and entire worlds with just a few simple voice commands or text inputs. Our pipeline is designed for ease of use and adoption. We've created an onboarding ramp that allows people to build on their phones and for any headset. We aim to be platform agnostic and build a robust framework that can integrate across services and platforms through an extension and overlay.

How we built it

Our platform is built on top of the popular A-Frame framework, which enables us to create 3D scenes and objects with ease. With our AI-powered interface, users can create shapes, textures, and even entire worlds with just a few simple voice commands or text inputs. The A-Frame prompt generator helps users achieve their 3D creation goals using the A-Frame framework. The prompts allow users to provide natural language feedback on the generated code, which is then used to refine the prompt and generate new code that creates the 3D scene.

Challenges we faced

Designing prompts that generate other prompts that move across APIs leaves lots of room for misinterpretation!

Accomplishments we're proud of

We've developed a full end-to-end pipeline where the user provides semantic input and generates a 3D shape. We're proud to have had a pineapple appear in 3D for the first time!

What we learned

AI amplifies human stupidity as much as it leverages existing intelligence. Wagyo taught us the importance of establishing a robust feedback loop between designer, AI, and user to develop a grounded understanding of the paired capabilities of AI and VR for building.

What's next for Wagyo

We plan to add more complex geometries, to create a more robust voice interface for use with headsets, and to integrate other generative AI models for textures, grounds, the sky, voices, characters, music, and animations.

Built With

• aframe
• openai