TexEditor: Structure-Preserving Text-Driven Texture Editing

Bo Zhao; Yihang Liu; Chenfeng Zhang; Huan Yang; Kun Gai; Wei Ji

cs.CVMar 19, 2026

TexEditor: Structure-Preserving Text-Driven Texture Editing

Authors: Bo Zhao, Yihang Liu, Chenfeng Zhang, Huan Yang, Kun Gai, Wei Ji

Organizations: Nanjing University · Shan Dong University · Zhejiang University · Kolors Team, Kuaishou Technology

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.

Figures & tables

Figure 1: Existing image editing models often struggle to preserve structural consistency with the original image during texture editing, and instead tend to regenerate the edited subject.

Figure 2: The training pipeline of TexEditor . (a) The model undergoes Supervised Fine-Tuning (SFT) on a synthetic dataset rendering 3D assets in Blender. (b) It is then optimized via Reinforcement Learning (RL), utilizing a multi-faceted reward system that incorporates Gemini for instruction alignment and structural analysis to balance texture editing with structural preservation.

Figure 3: Generation of image-instruction pairs in TexBlender . We first identify target object groups within filtered 3D scenes. Visual variations are created via (b1) Attribute Editing (adjusting shader parameters) or (b2) Texture Replacement (applying MatSynth textures). Finally, (b3) an Instruction Refiner leverages visual cues (difference images) and segmentation masks (SAM3) to generate precise, grounded text instructions that accurately describe the texture changes.

Figure 4: Visualization of different structure extractors and distance metrics in structure consistency loss

Name	Protocol	Task	Data Source	Scene	Size	Quality Control
Alchemist	Not released	Attribute modification	Blender (synthetic)	Single-object	150	Not specified
TexBench	Open-source	Attribute + replacement	COCO (real images)	Real-world	825	Manual inspection

Table 1: Comparison with existing texture editing benchmarks.

Figure 5: Qualitative Comparison of Different Evaluation Metrics. Results favored by Gemini, structural, and TexEval scores are shown in green, blue, and red; check marks indicate human preference. TexEval balances the structure and semantics and aligns best with humans.

Figure 6: Qualitative Comparison. TexEditor achieves the best visual results by modifying texture according to instructions while preserving the target object’s structure.

Method	Texture			Attribute
Method	Inst. $\uparrow$	Structure $\uparrow$	TexEval $\uparrow$	Inst. $\uparrow$	Structure $\uparrow$	TexEval $\uparrow$
Qwen-2509	0.767	0.642	0.717	0.584	0.408	0.514
ALchemist	0.761	0.711	0.741	0.631	0.512	0.583
TexEditor-Base	0.796	0.733	0.767	0.670	0.544	0.620
Qwen-2511	0.789	0.569	0.701	0.719	0.362	0.576
Nano Banana	0.726	0.584	0.669	0.530	0.332	0.451

Table 2: Quantitative comparison cross texture and attribute editing tasks within the TexBench

Method	Texture			Attribute
Method	Inst. $\uparrow$	Edge $\uparrow$	TexEval $\uparrow$	Inst. $\uparrow$	Edge $\uparrow$	TexEval $\uparrow$
Qwen-2509	0.791	0.841	0.811	0.777	0.856	0.809
Nano Banana	0.855	0.497	0.712	0.782	0.588	0.704
Nano Banana Pro	0.923	0.752	0.855	0.865	0.883	0.872
TexEditor	0.961	0.988	0.972	0.965	0.982	0.972

Table 3: Performance comparison cross texture and attribute editing tasks within the Blender-based benchmark

Config	SFT Data			RL Loss				Texture			Attribute
Config	None	Simple	Ours	None	Instrution	Structure	Norm. Structure	Instrution $\uparrow$	Structure $\uparrow$	TexEval $\uparrow$	Instrution $\uparrow$	Structure $\uparrow$	TexEval $\uparrow$
a	$\checkmark$			$\checkmark$				0.767	0.642	0.717	0.584	0.408	0.514
b		$\checkmark$		$\checkmark$				0.761	0.711	0.741	0.631	0.512	0.583
c			$\checkmark$	$\checkmark$				0.796	0.723	0.767	0.670	0.544	0.620
d	$\checkmark$				$\checkmark$		$\checkmark$	0.801	0.823	0.801	0.628	0.717	0.663
e		$\checkmark$			$\checkmark$		$\checkmark$	0.822	0.845	0.831	0.683	0.771	0.718

Table 4: Ablation study on different combinations of SFT data and RL loss.

Appendix figures & tables9 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 7: Qualitative comparisons on the attribute editing subtask of TexBench

Figure 8: Qualitative comparisons on the texture replacement subtask of TexBench. We additionally include ZeST, an image-to-image based texture transfer method, for comparison alongside instruction-based editing models.

Figure 9: Qualitative comparisons on Blender-based benchmarks.

Figure 10: Qualitative comparisons on texture-related subtasks of the ImgEdit benchmark

Method	Adjust $\uparrow$	Background $\uparrow$	Compose $\uparrow$	Action $\uparrow$	Add $\uparrow$
Qwen-2509	4.05	3.87	2.17	3.33	4.47
TexEditor (Ours)	4.13	4.13	2.39	2.79	4.28

Appendix

Table 5: Scores across different edit types in the ImgEdit benchmark.

Figure 12: Pipeline for TexBench Data and RL Image–Instruction Pair Construction

Figure 13: Screenshot of the web-based human evaluation interface

Figure 14: TexEval Performance under Different

\alpha

Settings

Image Editing

Image editing changes selected visual content while preserving the parts of an image that should remain unchanged.

See all 315 papers

Rethinking Structure Preservation in Text-Guided Image Editing with Visual Autoregressive Models
Mar 30, 2026Tao Xia, Jiawei Liu, Yukun Zhang +3Text-Guided Image Editing Image Editing
Beijing Institute of Technology, Beijing, China · Shenyang Institute of Automation, CAS, Shenyang, China · Meitu Inc, MTLab, Beijing, China
TextSculptor: Training and Benchmarking Scene Text Editing
May 20, 2026Yiheng Lin, Siyu Jiao, Xiaohan Lan +12Synthetic Data Augmentation Text-Guided Image Editing
Beijing Jiaotong University · Bytedance
STEPS: Scene Text Editing with Preserved Style Using Diffusion and Contrastive Style Encoding
Sep 29, 2026Nicolas Thiebaut, Nameer Hirschkind, Xiao Yu +1Text-Guided Image Editing Diffusion-Based Image Editing
Roblox San Mateo, CA, USA

Mar 30, 2026J/K move · Enter open · S save

Rethinking Structure Preservation in Text-Guided Image Editing with Visual Autoregressive Models

Tao Xia, Jiawei Liu, Yukun Zhang, Ting Liu+2

Beijing Institute of Technology, Beijing, China · Shenyang Institute of Automation, CAS, Shenyang, China · Meitu Inc, MTLab, Beijing, China

Text-Guided Image Editing Image Editing Visual Autoregressive Models

Open paper AskPDF arXiv

Visual autoregressive (VAR) models have recently emerged as a promising family of generative models, enabling a wide range of downstream vision tasks such as text-guided image editing. By shifting the editing paradigm from noise manipulation in diffusion-based methods to token-level operations, VAR-based approaches achieve better background preservation and significantly faster inference. However, existing VAR-based editing methods still face two key challenges: accurately localizing editable tokens and maintaining structural consistency in the edited results. In this work, we propose a novel text-guided image editing framework rooted in an analysis of intermediate feature distributions within VAR models. First, we introduce a coarse-to-fine token localization strategy that can refine editable regions, balancing editing fidelity and background preservation. Second, we analyze the intermediate representations of VAR models and identify structure-related features, by which we design a simple yet effective feature injection mechanism to enhance structural consistency between the edited and source images. Third, we develop a reinforcement learning-based adaptive feature injection scheme that automatically learns scale- and layer-specific injection ratios to jointly optimize editing fidelity and structure preservation. Extensive experiments demonstrate that our method achieves superior structural consistency and editing quality compared with state-of-the-art approaches, across both local and global editing scenarios.

Abstract

Image Editing

Explore similar work

Rethinking Structure Preservation in Text-Guided Image Editing with Visual Autoregressive Models

TextSculptor: Training and Benchmarking Scene Text Editing

STEPS: Scene Text Editing with Preserved Style Using Diffusion and Contrastive Style Encoding