Map2World: Segment Map Conditioned Text to 3D World Generation

TL;DR

Map2World is a novel framework for 3D world generation conditioned on user-defined segment maps, ensuring global consistency and detailed, controllable environments with strong generalization capabilities.

Contribution

The paper introduces Map2World, a new method that enables flexible, consistent 3D world generation conditioned on arbitrary segment maps, with a detail enhancer for fine details.

Findings

Outperforms existing methods in user-controllability and scale consistency.

Enables generation of complex 3D worlds with fine-grained details.

Demonstrates robust generalization across diverse domains.

Abstract

3D world generation is essential for applications such as immersive content creation or autonomous driving simulation. Recent advances in 3D world generation have shown promising results; however, these methods are constrained by grid layouts and suffer from inconsistencies in object scale throughout the entire world. In this work, we introduce a novel framework, Map2World, that first enables 3D world generation conditioned on user-defined segment maps of arbitrary shapes and scales, ensuring global-scale consistency and flexibility across expansive environments. To further enhance the quality, we propose a detail enhancer network that generates fine details of the world. The detail enhancer enables the addition of fine-grained details without compromising overall scene coherence by incorporating global structure information. We design the entire pipeline to leverage strong priors from asset generators, achieving robust generalization across diverse domains, even under limited training data for scene generation. Extensive experiments demonstrate that our method significantly outperforms existing approaches in user-controllability, scale consistency, and content coherence, enabling users to generate 3D worlds under more complex conditions.