Style-NeRF2NeRF: 3D Style Transfer From Style-Aligned Multi-View Images

TL;DR

This paper introduces a pipeline for 3D style transfer that uses multi-view images generated by diffusion models to refine a NeRF scene, enabling diverse artistic style application with preview capabilities.

Contribution

The method uniquely combines multi-view diffusion-based image generation with NeRF refinement for effective 3D style transfer, allowing flexible prompt testing and high-quality stylization.

Findings

Effective transfer of diverse artistic styles to 3D scenes

Allows preview of stylized results before NeRF fine-tuning

Achieves competitive quality in 3D style transfer

Abstract

We propose a simple yet effective pipeline for stylizing a 3D scene, harnessing the power of 2D image diffusion models. Given a NeRF model reconstructed from a set of multi-view images, we perform 3D style transfer by refining the source NeRF model using stylized images generated by a style-aligned image-to-image diffusion model. Given a target style prompt, we first generate perceptually similar multi-view images by leveraging a depth-conditioned diffusion model with an attention-sharing mechanism. Next, based on the stylized multi-view images, we propose to guide the style transfer process with the sliced Wasserstein loss based on the feature maps extracted from a pre-trained CNN model. Our pipeline consists of decoupled steps, allowing users to test various prompt ideas and preview the stylized 3D result before proceeding to the NeRF fine-tuning stage. We demonstrate that our method can transfer diverse artistic styles to real-world 3D scenes with competitive quality. Result videos are also available on our project page: https://haruolabs.github.io/style-n2n/