SynCity 3000: Bootstrapping Scene-Scale 3D Diffusion
2026-07-06 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors created SynCity 3000, a system that can generate big 3D scenes that make sense overall and let users control the layout closely. They improved current methods that make 3D objects from single images so they can build entire scenes by turning the generator into a tool that works like a convolution (a common operation in image processing). To train this tool, they made a new synthetic data engine to produce scene-like training data, fixing the problem of not having enough real 3D scene data. Their system takes a special image representing the whole scene and produces detailed 3D environments of any size based on user descriptions.
3D scene generationimage-to-3D generatorconvolutional operatorsynthetic data enginefine-tuningdimetric imagelayout controlscene coherenceprompt-based generation3D assets
Authors
Paul Engstler, Iro Laina, Christian Rupprecht, Andrea Vedaldi
Abstract
We present SynCity 3000, a framework for generating 3D scenes that are globally coherent while enabling fine-grained layout control. Building on the ability of current image-to-3D generators to produce complex 3D assets from a single image, we extend this capability to the scale of entire scenes by adapting the generator to be applicable as a convolutional operator. We achieve this by fine-tuning the model on scene-like data generated by a new synthetic data engine, which we propose to address the scarcity of 3D scene data for training. The convolutional generator is then applied to a dimetric image of the entire scene, generated from the user prompt, resulting in 3D scenes of arbitrary size and complexity. Across diverse prompts and layouts, SynCity 3000 produces large, coherent, and detailed scenes, addressing the shortcomings of prior approaches to 3D scene generation.