Published February 28, 2026 | Version v1
Model Open

AnyStyle

Authors/Creators

  • 1. ROR icon Warsaw University of Technology
  • 2. ROR icon Sano – Centre for Computational Personalised Medicine International Research Foundation
  • 3. ROR icon Jagiellonian University

Description

Given a set of unposed input images and a style conditioning signal provided as either text or an image, our method generates a tylized 3D scene represented with 3D Gaussian Splats in a single forward pass. The reconstructed scene can be stylized in under 0.1 second per input content image.

Files

AnyStyle.zip

Files (13.1 GB)

Name Size Download all
md5:2af7050e4c7f0bfda657b00c2fb8c9fd
13.1 GB Preview Download

Additional details

Software

Repository URL
https://github.com/joaxkal/AnyStyle
Programming language
Python
Development Status
Wip