TexEditor: Structure-Preserving Text-Driven Texture Editing

TL;DR

TexEditor is a novel texture editing model that significantly improves structural consistency during text-guided texture modifications by leveraging new datasets and structure-preserving training methods.

Contribution

The paper introduces TexEditor, a structure-preserving texture editing framework utilizing a new dataset and RL-based approach to enhance structural fidelity in texture editing.

Findings

Outperforms existing models on Blender-based benchmarks

Achieves superior structural preservation in texture editing

Demonstrates strong generalization on ImgEdit benchmark

Abstract

Text-guided texture editing aims to modify object appearance while preserving the underlying geometric structure. However, our empirical analysis reveals that even SOTA editing models frequently struggle to maintain structural consistency during texture editing, despite the intended changes being purely appearance-related. Motivated by this observation, we jointly enhance structure preservation from both data and training perspectives, and build TexEditor, a dedicated texture editing model based on Qwen-Image-Edit-2509. Firstly, we construct TexBlender, a high-quality SFT dataset generated with Blender, which provides strong structural priors for a cold start. Sec- ondly, we introduce StructureNFT, a RL-based approach that integrates structure-preserving losses to transfer the structural priors learned during SFT to real-world scenes. Moreover, due to the limited realism and evaluation coverage of existing benchmarks, we introduce TexBench, a general-purpose real-world benchmark for text-guided texture editing. Extensive experiments on existing Blender-based texture benchmarks and our TexBench show that TexEditor consistently outperforms strong baselines such as Nano Banana Pro. In addition, we assess TexEditor on the general purpose benchmark ImgEdit to validate its generalization. Our code and data are available at https://github.com/KlingAIResearch/TexEditor.