Direct color and texture control improves 3D stylization quality
DiDE:Direct Injection with Color-Texture DEcoupling for 3D Stylization
Computer Vision and Pattern Recognition
Summary
It can be hard to change the look of 3D objects by separately adjusting their colors and textures. Existing methods usually change both together, making it difficult to get exactly the style you want. The authors found a way to separate color and texture adjustments in 3D images without needing extra training. Their approach processes content, color, and texture separately and merges them cleanly, keeping the original shape unchanged. Tests show their method transfers colors and textures more faithfully than previous techniques.
What this means in practice
- •For 3d artists and animators: Create 3D assets with independently adjustable color and texture to achieve precise artistic styles without retraining.
- •For game developers: Customize game character appearances by separately controlling color schemes and surface textures for better visual variety.
Authors
Tao Wu, Alexandra Gomez-Villa, Senmao Li, Yaxing Wang, Joost van de Weijer, Kai Wang
Abstract
Recent advances in rectified flow-based image-to-3D generative models have enabled high-fidelity 3D asset generation. Building on this, a growing line of work has exploited these strong 3D priors for training-free stylization, transferring visual attributes from a reference image onto a generated 3D asset. However, existing methods enforce an all-or-nothing paradigm: color and texture are transferred jointly, with no mechanism to control them independently -- a limitation we formalize as Disentangled 3D Stylization(Disen3D). To address this, we propose DiDE, the first training-free framework for Disen3D. Key to our approach is the observation that the structured latent space of image-to-3D models is overcomplete with respect to texture: texture information occupies only a small subset of the style-significant channels, leaving a free subspace available for independent color encoding. DiDE exploits this via a channel partition mechanism that processes a content image, a texture reference, and a color reference through dedicated branches and composes both style signals interference-free at every self-attention layer, preserving content geometry throughout. Experiments on Disen3D-Bench, our newly collected multi-reference benchmark, show that DiDE consistently outperforms 2D and 3D stylization baselines in color fidelity, texture transfer, and content preservation.