C3Editor: Achieving Controllable Consistency in 2D Model for 3D Editing

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Tao, Zeng, Ding, Zheng, Chen, Zeyuan, Zhang, Xiang, Li, Leizhi, Tu, Zhuowen
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915588144627712
author Tao, Zeng
Ding, Zheng
Chen, Zeyuan
Zhang, Xiang
Li, Leizhi
Tu, Zhuowen
author_facet Tao, Zeng
Ding, Zheng
Chen, Zeyuan
Zhang, Xiang
Li, Leizhi
Tu, Zhuowen
contents Existing 2D-lifting-based 3D editing methods often encounter challenges related to inconsistency, stemming from the lack of view-consistent 2D editing models and the difficulty of ensuring consistent editing across multiple views. To address these issues, we propose C3Editor, a controllable and consistent 2D-lifting-based 3D editing framework. Given an original 3D representation and a text-based editing prompt, our method selectively establishes a view-consistent 2D editing model to achieve superior 3D editing results. The process begins with the controlled selection of a ground truth (GT) view and its corresponding edited image as the optimization target, allowing for user-defined manual edits. Next, we fine-tune the 2D editing model within the GT view and across multiple views to align with the GT-edited image while ensuring multi-view consistency. To meet the distinct requirements of GT view fitting and multi-view consistency, we introduce separate LoRA modules for targeted fine-tuning. Our approach delivers more consistent and controllable 2D and 3D editing results than existing 2D-lifting-based methods, outperforming them in both qualitative and quantitative evaluations.
format Preprint
id arxiv_https___arxiv_org_abs_2510_04539
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle C3Editor: Achieving Controllable Consistency in 2D Model for 3D Editing
Tao, Zeng
Ding, Zheng
Chen, Zeyuan
Zhang, Xiang
Li, Leizhi
Tu, Zhuowen
Graphics
Computer Vision and Pattern Recognition
Existing 2D-lifting-based 3D editing methods often encounter challenges related to inconsistency, stemming from the lack of view-consistent 2D editing models and the difficulty of ensuring consistent editing across multiple views. To address these issues, we propose C3Editor, a controllable and consistent 2D-lifting-based 3D editing framework. Given an original 3D representation and a text-based editing prompt, our method selectively establishes a view-consistent 2D editing model to achieve superior 3D editing results. The process begins with the controlled selection of a ground truth (GT) view and its corresponding edited image as the optimization target, allowing for user-defined manual edits. Next, we fine-tune the 2D editing model within the GT view and across multiple views to align with the GT-edited image while ensuring multi-view consistency. To meet the distinct requirements of GT view fitting and multi-view consistency, we introduce separate LoRA modules for targeted fine-tuning. Our approach delivers more consistent and controllable 2D and 3D editing results than existing 2D-lifting-based methods, outperforming them in both qualitative and quantitative evaluations.
title C3Editor: Achieving Controllable Consistency in 2D Model for 3D Editing
topic Graphics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.04539