# Customize your NeRF: Adaptive Source Driven 3D Scene Editing via Local-Global Iterative Training

Source: [Customize your NeRF: Adaptive Source Driven 3D Scene Editing via Local-Global Iterative Training](<https://www.sota2.com/research/paper/68fa71571e209131cea50aa4>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-02T04:23:44.604Z

Authors: Runze He, Shaofei Huang, Xuecheng Nie, Tianrui Hui, Luoqi Liu, Jiao Dai, Jizhong Han, Guanbin Li, Si Liu

Published: 2023-12-04T06:25:06.000Z

## Abstract

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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Image-driven 3D Scene Editing on Image-driven editing set 5 scenes (test)](<https://www.sota2.com/research/sota/image-driven-3d-scene-editing-on-image-driven-editing-set-5-scenes-test.md>): Image-driven 3D Scene Editing; Image-driven editing set 5 scenes (test); CLIP Direction Similarity: 20.07.

- [Text-driven 3D scene editing on Text-driven editing evaluation set 10 edits across 2 scenes](<https://www.sota2.com/research/sota/text-driven-3d-scene-editing-on-text-driven-editing-evaluation-set-10-edits-across-2-scenes.md>): Text-driven 3D scene editing; Text-driven editing evaluation set 10 edits across 2 scenes; CLIP Directionality Score: 22.66.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2312.01663>)

[PDF](<https://arxiv.org/pdf/2312.01663.pdf>)

[Implementation](<https://customnerf.github.io/>)