DreamStyle3D: Efficient 3D Stylized Asset Generation via Dual-Attention Disentanglement
2026-07-27 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors created DreamStyle3D, a new way to quickly style 3D models that keeps their shapes accurate while adding cool artistic effects. Unlike previous methods that go from 2D images to 3D, their approach works directly in 3D by separating the style from the shape. They also made a big dataset to help train and test their method. Their system can produce high-quality, stylized 3D objects in about 10 seconds, improving speed without losing detail.
3D asset generationstylizationgeometry consistencycross-attention mechanismstyle injectiontraining data pipelinemodel generalizationcontent-style tripletsvirtual realitycomputer graphics
Authors
Kai Wang, Ziheng Ouyang, Xuying Zhang, Ming-Ming Cheng, Qibin Hou
Abstract
With the growth of gaming, animation, and virtual reality industries, the demand for efficient generation of stylized 3D assets is rapidly increasing. However, existing approaches still struggle to jointly preserve style fidelity, geometric consistency, and generation efficiency, as most of them still rely on indirect 2D-to-3D stylization pipelines. This motivates a native 3D stylization framework that can explicitly disentangle style from geometry while remaining efficient. To this end, we propose DreamStyle3D, an efficient framework for stylized 3D asset generation built on a Decoupled Dual Cross-Attention mechanism. Our method explicitly separates geometric and stylistic features to enable efficient style injection while preserving structural consistency, and further adopts a lightweight training strategy to enhance style consistency and model generalization. In addition, we build an automated data pipeline and construct a dataset of about 15K content-style-stylized triplets for training and evaluation. Extensive experiments demonstrate that our DreamStyle3D can generate high-fidelity, geometrically consistent stylized 3D assets within 10 seconds, substantially improving efficiency while maintaining superior style quality and offering a new solution for 3D content creation. The code and data are available at https://github.com/HVision-NKU/DreamStyle3D.