The hardest problem in AI made image to 3D models right now isn't the shape, it's the mesh underneath. From the time I launched this project, I've been trying to get the human OUT of the loop. While I work on that, here's a primer on topology! https://x.com/i/article/2113705880223113217