arXiv Computer Vision By Hanzhou Liu, Jia Huang, Mi Lu, Srikanth Saripalli, Peng Jiang

Stylos: Multi-View 3D Stylization with Single-Forward Gaussian Splatting

Read the original on arXiv Computer Vision →

Stylos is a single-forward 3D Gaussian framework that performs 3D style transfer from a single image to a multi-view collection without requiring per-scene optimization or precomputed poses. It uses a Transformer backbone with geometry and style pathways, employing self-attention for geometric fidelity and global cross-attention for visual consistency, and introduces a voxel-based 3D style loss to align scene features with style statistics. Experiments on multiple datasets show that Stylos achieves high-quality, zero-shot stylization that generalizes to unseen categories, scenes, and styles.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

arXiv Computer Vision
Oct 7

InStyle: Instant Appearance Stylization of 3D Shapes

The paper introduces GaussianBlender, a feed‑forward framework that enables instant, high‑fidelity, geometry‑preserving, and multi‑view consistent 3D stylization driven by text. It learns disentangled latent spaces for geometry and appearance using spatially‑grouped 3D Gaussians, and applies a latent diffusion model for text‑conditioned edits. Evaluations demonstrate that GaussianBlender outperforms methods requiring per‑instance optimization, making scalable 3D stylization practical for large‑scale production.

By Melis Ocal, Xiaoyan Xing, Yue Li, Ngo Anh Vien, Sezer Karaoglu, Theo Gevers
arXiv Computer Vision
Sep 18

ArtNVG: Content-Style Separated Artistic Neighboring-View Gaussian Stylization

ArtNVG is a 3D stylization framework that uses reference style images to produce stylized 3D scenes with high reconstruction quality. Built on 3D Gaussian Splatting, it introduces Content‑Style Separated Control and Attention‑based Neighboring‑View Alignment to maintain local color and texture consistency across views. Experiments show that ArtNVG outperforms existing methods in content preservation, style alignment, and local consistency.

By Zixiao Gu, Mengtian Li, Ruhua Chen, Zhongxia Ji, Sichen Guo, Zhenye Zhang, Guangnan Ye, Zuo Hu
arXiv Computer Vision
Sep 1

DReSG: Diffusion Residuals for Stylized Gaussian Splatting

DReSG introduces a 3D-grounded residual-feedback framework for stylizing scenes represented by 3D Gaussian Splatting. It uses attention-guided diffusion proposals as residual targets relative to current renders and progressively integrates these residuals into a shared Gaussian scene via multi-view feedback. The method stabilizes and controls the feedback by modulating residual strength, selecting coverage-aware views, and filtering color updates, achieving competitive stylization while preserving scene structure and cross-view stability.

By Zhongliang Liu, Wenjie Liu, Yang Li
arXiv Computer Vision
Oct 2

DiDE:Direct Injection with Color-Texture DEcoupling for 3D Stylization

DiDE is a training‑free framework that enables independent control of color and texture when stylizing 3D assets generated by rectified flow‑based image‑to‑3D models. It leverages the overcomplete latent space of these models, partitioning channels so that texture occupies a small subset while color is encoded in a free subspace. Experiments on the newly collected Disen3D‑Bench benchmark demonstrate that DiDE outperforms existing 2D and 3D stylization methods in color fidelity, texture transfer, and content preservation.

By Tao Wu, Alexandra Gomez-Villa, Senmao Li, Yaxing Wang, Joost van de Weijer, Kai Wang