Customize your NeRF: Adaptive Source Driven 3D Scene Editing via Local-Global Iterative Training
This work addresses the problem of 3D scene editing for users needing to modify scenes based on text or image prompts, with incremental improvements in handling foreground accuracy and consistency.
The paper tackles the problem of adaptive source-driven 3D scene editing using NeRF, addressing challenges in foreground-only manipulation and multi-view consistency, and shows that their CustomNeRF model produces precise editing results in real scenes for both text- and image-driven settings.
In this paper, we target the adaptive source driven 3D scene editing task by proposing a CustomNeRF model that unifies a text description or a reference image as the editing prompt. However, obtaining desired editing results conformed with the editing prompt is nontrivial since there exist two significant challenges, including accurate editing of only foreground regions and multi-view consistency given a single-view reference image. To tackle the first challenge, we propose a Local-Global Iterative Editing (LGIE) training scheme that alternates between foreground region editing and full-image editing, aimed at foreground-only manipulation while preserving the background. For the second challenge, we also design a class-guided regularization that exploits class priors within the generation model to alleviate the inconsistency problem among different views in image-driven editing. Extensive experiments show that our CustomNeRF produces precise editing results under various real scenes for both text- and image-driven settings.