Abstract
In 3D modeling, designers often use an existing 3D model as a reference to create new ones. This practice has inspired the development of Phidias, a novel generative model that uses diffusion for reference-augmented 3D generation. Given an image, our method leverages a retrieved or user-provided 3D reference model to guide the generation process, thereby enhancing the generation quality, generalization ability, and controllability. Our model integrates three key components: 1) meta-ControlNet that dynamically modulates the conditioning strength, 2) dynamic reference routing that mitigates misalignment between the input image and 3D reference, and 3) self-reference augmentations that enable self-supervised training with a progressive curriculum. Collectively, these designs result in a clear improvement over existing methods. Phidias establishes a unified framework for 3D generation using text, image, and 3D conditions with versatile applications.
Paper: arxiv.org/abs/2409.11406
Code: github.com/3DTopia/Phidias-Diffusion
Models: huggingface.co/ZhenweiWang/…/main
Project Page: rag-3d.github.io
DoucheBagMcSwag@lemmy.dbzer0.com 4 weeks ago
Ah fuck even the CG animation industry is cooked
Even_Adder@lemmy.dbzer0.com 4 weeks ago
There’s talk of constant shortages of people to work on CG shots for movies. I think that this kind of stuff will just result in induced demand for work.