CHOrD: Generation of Collision-Free, House-Scale, and Organized Digital Twins for 3D Indoor Scenes with Controllable Floor Plans and Optimal Layouts
Chong Su, Yingbin Fu, Zheyuan Hu, Jing Yang, Param Hanji, Shaojun Wang, Xuan Zhao, Cengiz Öztireli, Fangcheng Zhong
2025-03-17
Summary
This paper introduces CHOrD, a new system that automatically creates realistic 3D models of indoor spaces, like houses, that are organized and free of collisions.
What's the problem?
Creating accurate and realistic 3D models of indoor spaces is challenging. Existing methods often result in collisions between objects or lack a structured organization, and they don't always work well with different floor plans.
What's the solution?
CHOrD uses a 2D image-based representation as an intermediate step, which helps prevent collisions by identifying them as unusual scenarios during the creation process. It can also handle complex floor plans and create coherent layouts for entire houses.
Why it matters?
This work matters because it provides a more reliable and versatile way to generate realistic 3D indoor environments, which can be useful for applications like virtual reality, architectural design, and creating virtual tours of homes.
Abstract
We introduce CHOrD, a novel framework for scalable synthesis of 3D indoor scenes, designed to create house-scale, collision-free, and hierarchically structured indoor digital twins. In contrast to existing methods that directly synthesize the scene layout as a scene graph or object list, CHOrD incorporates a 2D image-based intermediate layout representation, enabling effective prevention of collision artifacts by successfully capturing them as out-of-distribution (OOD) scenarios during generation. Furthermore, unlike existing methods, CHOrD is capable of generating scene layouts that adhere to complex floor plans with multi-modal controls, enabling the creation of coherent, house-wide layouts robust to both geometric and semantic variations in room structures. Additionally, we propose a novel dataset with expanded coverage of household items and room configurations, as well as significantly improved data quality. CHOrD demonstrates state-of-the-art performance on both the 3D-FRONT and our proposed datasets, delivering photorealistic, spatially coherent indoor scene synthesis adaptable to arbitrary floor plan variations.