Controllable Text-to-3D Generation via Surface-Aligned Gaussian Splatting

arXiv:2403.09981v1 Announce Type: new
Abstract: While text-to-3D and image-to-3D generation tasks have received considerable attention, one important but under-explored field between them is controllable text-to-3D generation, which we mainly focus on in this work. To address this task, 1) we introduce Multi-view ControlNet (MVControl), a novel neural network architecture designed to enhance existing pre-trained multi-view diffusion models by integrating additional input conditions, such as edge, depth, normal, and scribble maps. Our innovation lies in the introduction of a conditioning module …

abstract architecture arxiv attention controlnet cs.cv focus image network network architecture neural network novel surface tasks text them type via view work