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

By Lukas Höllein, Ang Cao, Andrew Owens, Justin Johnson, Matthias Nießner. IEEE International Conference on Computer Vision, 2023

308 citations recorded by the source corpus.

We present Text2Room 1, 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 [56], [41] or zoom-out trajectories [18] 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.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/95aa6fa4e42387561cff22378348d528adea37f2

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

CAT3D: Create Anything in 3D with Multi-View Diffusion Models (2024, methodology): “Incorporating priors such as monocular depth models or inpainting models has been useful for creating 3D scenes, but tends to result in poor global geometry [32, 33, 34, 35].”
FSGS: Real-Time Few-shot View Synthesis using Gaussian Splatting (2023, background): “Tremendous works focus on improving its efficiency [7, 15, 17, 26, 31, 38, 43], quality [1, 3, 4, 9, 20, 42, 48, 52], generalizing to unseen scenes [8, 10, 24, 44, 53, 59], applying artistic effects [14, 22, 49, 60] and 3D genera-2 tion [5, 6, 18, 21, 25, 27, 28, 34, 41, 45].”
ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation (2023, methodology): “The integration of off-the-shelf depth estimation models [37], as utilized in other studies [16, 58], could potentially enhance the accuracy and detail of the scene generation process.”
Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors (2023, methodology): “The developed methods improved the underlying 3D representation [43, 11, 84] and 3D consistency of the supervision [46, 77]; explored task-specific priors [28, 33, 70] and additional controls [54].”