DreamScene：基于3D高斯分布的端到端文本到3D场景生成

摘要

从自然语言生成三维场景在游戏、电影和设计领域展现出巨大潜力。然而，现有方法在自动化、三维一致性及细粒度控制方面仍面临挑战。我们提出了DreamScene，一个端到端的框架，旨在从文本或对话中生成高质量且可编辑的三维场景。DreamScene首先通过场景规划模块，利用GPT-4智能体推断物体语义与空间约束，构建混合图。随后，基于图的布局算法生成结构化且无碰撞的场景布局。在此布局基础上，形态模式采样（FPS）采用多时间步采样与重建优化技术，快速生成逼真的物体几何形态。为确保全局一致性，DreamScene采用了渐进式相机采样策略，适应室内外多种场景需求。最后，系统支持细粒度的场景编辑功能，包括物体移动、外观调整及四维动态运动。实验表明，DreamScene在质量、一致性和灵活性上均超越现有方法，为开放域三维内容创作提供了实用解决方案。代码与演示详见https://jahnsonblack.github.io/DreamScene-Full/。

English

Generating 3D scenes from natural language holds great promise for applications in gaming, film, and design. However, existing methods struggle with automation, 3D consistency, and fine-grained control. We present DreamScene, an end-to-end framework for high-quality and editable 3D scene generation from text or dialogue. DreamScene begins with a scene planning module, where a GPT-4 agent infers object semantics and spatial constraints to construct a hybrid graph. A graph-based placement algorithm then produces a structured, collision-free layout. Based on this layout, Formation Pattern Sampling (FPS) generates object geometry using multi-timestep sampling and reconstructive optimization, enabling fast and realistic synthesis. To ensure global consistent, DreamScene employs a progressive camera sampling strategy tailored to both indoor and outdoor settings. Finally, the system supports fine-grained scene editing, including object movement, appearance changes, and 4D dynamic motion. Experiments demonstrate that DreamScene surpasses prior methods in quality, consistency, and flexibility, offering a practical solution for open-domain 3D content creation. Code and demos are available at https://jahnsonblack.github.io/DreamScene-Full/.

DreamScene：基于3D高斯分布的端到端文本到3D场景生成

DreamScene: 3D Gaussian-based End-to-end Text-to-3D Scene Generation

摘要

Support