Text2Room: Extracting Textured 3D Meshes from 2D Text-to-Image Models

TL;DR

Text2Room is a novel method that generates complete, textured 3D room-scale meshes from text prompts by synthesizing images, estimating depth, and fusing views into a seamless 3D scene.

Contribution

It introduces a new approach that creates full 3D scenes with multiple objects and textures from text, combining pre-trained models and a continuous alignment strategy.

Findings

First to generate room-scale textured 3D scenes from text

Produces seamless, multi-object 3D meshes with explicit geometry

Outperforms existing methods in qualitative and quantitative metrics

Abstract

We present Text2Room, a method for generating room-scale textured 3D meshes from a given text prompt as input. To this end, we leverage pre-trained 2D text-to-image models to synthesize a sequence of images from different poses. In order to lift these outputs into a consistent 3D scene representation, we combine monocular depth estimation with a text-conditioned inpainting model. The core idea of our approach is a tailored viewpoint selection such that the content of each image can be fused into a seamless, textured 3D mesh. More specifically, we propose a continuous alignment strategy that iteratively fuses scene frames with the existing geometry to create a seamless mesh. Unlike existing works that focus on generating single objects or zoom-out trajectories from text, our method generates complete 3D scenes with multiple objects and explicit 3D geometry. We evaluate our approach using qualitative and quantitative metrics, demonstrating it as the first method to generate room-scale 3D geometry with compelling textures from only text as input.