DReSG: Diffusion Residuals for Stylized Gaussian Splatting
Computer Vision and Pattern RecognitionGraphics
Summary
The authors introduce DReSG, a new method to style 3D scenes made of Gaussian splats by using a special feedback system that gradually updates the scene based on differences from a reference style. Unlike other methods that can cause inconsistent views or unwanted artifacts, their approach carefully controls updates and chooses views to keep the scene stable and consistent. Experiments show DReSG can create stylized 3D scenes that keep the original structure and look good from multiple angles. This helps in making 3D content creation faster and more controllable.
Authors
Zhongliang Liu, Wenjie Liu, Yang Li
Abstract
Reference-guided stylization of scenes represented by 3D Gaussian Splatting (3DGS) is important for efficient and controllable 3D content creation. Existing VGG-feature-based 3D stylization methods provide stable rendered-view optimization, but often under-represent expressive reference style cues; diffusion models offer stronger image priors, yet direct per-view or score-based diffusion guidance can lead to view drift, local artifacts, and hard-to-control appearance updates. We present DReSG, a 3D-grounded residual-feedback framework for stylized Gaussian splatting. DReSG represents attention-guided diffusion proposals as residual targets relative to the current render, and progressively absorbs these residuals into a shared Gaussian scene through multi-view Gaussian feedback. To make this feedback stable and controllable, DReSG modulates residual strength during target construction and combines coverage-aware view selection with conflict-filtered color updates during multi-view fitting. Extensive experiments demonstrate that DReSG achieves competitive reference-guided stylization while better preserving scene structure and cross-view stability. Our project page is available at https://vpx-ecnu.github.io/DReSG-website/.