Editing 3D Scenes via Text Prompts without Retraining

TL;DR

This paper introduces DN2N, a text-driven 3D scene editing method that enables universal editing capabilities without retraining, using off-the-shelf 2D image editing models and a filtering process to maintain 3D consistency.

Contribution

The paper presents a novel approach that allows direct, retraining-free editing of 3D scenes via text prompts, leveraging cross-view regularization and noise removal techniques.

Findings

Enables multiple editing types like appearance, weather, material, and style transfer.

Achieves generalization across different scenes without scene-specific retraining.

Maintains 3D consistency during editing through filtering and regularization.

Abstract

Numerous diffusion models have recently been applied to image synthesis and editing. However, editing 3D scenes is still in its early stages. It poses various challenges, such as the requirement to design specific methods for different editing types, retraining new models for various 3D scenes, and the absence of convenient human interaction during editing. To tackle these issues, we introduce a text-driven editing method, termed DN2N, which allows for the direct acquisition of a NeRF model with universal editing capabilities, eliminating the requirement for retraining. Our method employs off-the-shelf text-based editing models of 2D images to modify the 3D scene images, followed by a filtering process to discard poorly edited images that disrupt 3D consistency. We then consider the remaining inconsistency as a problem of removing noise perturbation, which can be solved by generating training data with similar perturbation characteristics for training. We further propose cross-view regularization terms to help the generalized NeRF model mitigate these perturbations. Our text-driven method allows users to edit a 3D scene with their desired description, which is more friendly, intuitive, and practical than prior works. Empirical results show that our method achieves multiple editing types, including but not limited to appearance editing, weather transition, material changing, and style transfer. Most importantly, our method generalizes well with editing abilities shared among a set of model parameters without requiring a customized editing model for some specific scenes, thus inferring novel views with editing effects directly from user input. The project website is available at https://sk-fun.fun/DN2N