Text2Mesh: Text-Driven Neural Stylization for Meshes
This provides a novel tool for 3D artists and designers to stylize meshes based on text descriptions, though it is incremental in combining CLIP with neural fields for 3D editing.
The paper tackles the problem of editing 3D mesh styles using text prompts, achieving intuitive controls by predicting color and geometric details without requiring pre-trained models or specialized datasets.
In this work, we develop intuitive controls for editing the style of 3D objects. Our framework, Text2Mesh, stylizes a 3D mesh by predicting color and local geometric details which conform to a target text prompt. We consider a disentangled representation of a 3D object using a fixed mesh input (content) coupled with a learned neural network, which we term neural style field network. In order to modify style, we obtain a similarity score between a text prompt (describing style) and a stylized mesh by harnessing the representational power of CLIP. Text2Mesh requires neither a pre-trained generative model nor a specialized 3D mesh dataset. It can handle low-quality meshes (non-manifold, boundaries, etc.) with arbitrary genus, and does not require UV parameterization. We demonstrate the ability of our technique to synthesize a myriad of styles over a wide variety of 3D meshes.