AI Diffusion Research
An exploration of image and 3D content generation with pretrained diffusion models.
Screenshot placeholder
Overview
These experiments used pretrained models to generate images and basic 3D content, with Stable Diffusion for 2D work and Shap-E for 3D.
The Problem
I wanted to understand how pretrained generative models could support workflows for different kinds of content.
My Contribution
Through The Knowledge Society, I worked with Stable Diffusion and Shap-E in Google Colab to explore image and 3D generation.
Development
I worked with pretrained models. Stable Diffusion supported image experiments, while Shap-E gave me a way to explore basic 3D generation.
Working across 2D and 3D meant learning separate workflows and examining different kinds of outputs.
Results
Explored 2D images and basic 3D content generation.