Preserving Subject-Clarity in Image Outpainting with Multiscale Wavelet Supervision

Abhilash Neog; Taewan Kim; Yi Wu; Xu Chen; Jian Jiao

cs.CVSep 5, 2026

Preserving Subject-Clarity in Image Outpainting with Multiscale Wavelet Supervision

Authors: Abhilash Neog, Taewan Kim, Yi Wu, Xu Chen, Jian Jiao

Organizations: Microsoft · Virginia Tech

Abstract

Commercial and advertising images are frequently affected by poor framing, partially cropped subjects, truncated text or logos, and insufficient context, all of which can reduce subject clarity, i.e., the ability of an image to clearly communicate its primary subject. Image outpainting offers a scalable solution by extending image boundaries and recovering missing content and context. However, existing diffusion-based outpainting methods often produce visually plausible completions while degrading subject fidelity through structural inconsistencies, semantic drift, or loss of fine-grained detail. To address this limitation, we propose a subject clarity outpainting framework that combines vision-language model (VLM)-guided semantic conditioning with multiscale wavelet supervision for subject-localized detail preservation. To support training, we develop a subject-centric data curation pipeline that constructs subject-intersecting outpainting pairs from advertising and natural images. The resulting objective introduces no additional inference cost and is designed to be compatible with diffusion-based backbones. Across four advertising and natural-image benchmarks, our method improves subject clarity, reducing subject-centered DreamSim error and FID on average by 3.0% and 2.4% over matched supervised fine-tuning, and by 10.8% and 7.7% over the strongest state-of-the-art approach per dataset, respectively.

Figures & tables

Panel of Figure 1 — Figure 1: What is subject clarity outpainting? General outpainting extends an image beyond its boundaries to synthesize missing scene content. We focus on the regime in which the original framing partially crops the primary subject. In this setting, outpainting must reconstruct the missing subject region while preserving subject fidelity —identity, structure, fine-grained appearance, and continuity with the visible portion. Existing methods can generate plausible surroundings yet distort the continued subject.

Figure 2: Overview of the proposed training framework. vision–language model provides scene guidance to the outpainting backbone. During training, the predicted and ground-truth images are compared in a multilevel wavelet space within the affected subject region. Subject-frequency statistics determine the relative importance of wavelet subbands, while the noise level controls the active scales. The resulting subject-aware wavelet loss complements the standard flow objective and is not used during inference.

Method	Backbone	Synth-Ads					Real-Ads
		Overall		Subject Clarity			Overall		Subject Clarity
		FID $\downarrow$	LPIPS full $\downarrow$	LPIPS Ω $\downarrow$	DINOv2 Ω $\downarrow$	DreamSim ctx $\downarrow$	FID $\downarrow$	LPIPS full $\downarrow$	LPIPS Ω $\downarrow$	DINOv2 Ω $\downarrow$	DreamSim ctx $\downarrow$
BrushNet [ 9 ]	SDXL	31.41	0.2590	0.5015	0.4586	0.0743	74.90	0.2839	0.5528	0.5124	0.1096
PowerPaint [ 32 ]	SD 1.5	31.79	0.2590	0.4152	0.3789	0.0815	72.77	0.2741	0.4799	0.4453	0.1184
VIP [ 29 ]	SD 1.5	25.26	0.2118	0.3957	0.3490	0.0514	55.20	0.2113	0.4064	0.3593	0.0783
FLUX.1 Fill [ 11 ]	FLUX 1.0	29.56	0.2327	0.3542	0.3157	0.0644	86.92	0.3148	0.4545	0.4083	0.1555

Table 1: Outpainting results on the Synth-Ads and Real-Ads test sets. Per-dataset best and second best are bold and underlined , respectively.

Method	Backbone	LVIS					Open Images
		Overall		Subject Clarity			Overall		Subject Clarity
		FID $\downarrow$	LPIPS full $\downarrow$	LPIPS Ω $\downarrow$	DINOv2 Ω $\downarrow$	DreamSim ctx $\downarrow$	FID $\downarrow$	LPIPS full $\downarrow$	LPIPS Ω $\downarrow$	DINOv2 Ω $\downarrow$	DreamSim ctx $\downarrow$
BrushNet [ 9 ]	SDXL	19.91	0.2587	0.5173	0.4910	0.0991	60.41	0.2674	0.5159	0.5100	0.1102
PowerPaint [ 32 ]	SD 1.5	22.49	0.2788	0.4530	0.4262	0.1065	66.27	0.2853	0.4588	0.4520	0.1193
VIP [ 29 ]	SD 1.5	15.82	0.2082	0.4107	0.3615	0.0708	48.18	0.2116	0.4062	0.3628	0.0756
FLUX.1 Fill [ 11 ]	FLUX 1.0	22.76	0.2594	0.3953	0.3510	0.0951	63.10	0.2716	0.3929	0.3686	0.1020

Table 2: Outpainting results on the LVIS and Open Images test sets. Per-dataset best and second best are bold and underlined , respectively.

Panel of Figure 3 — Figure 3: The Input / GT column shows the model input and ground-truth outpaint region in gray, with the black line marking the outpaint boundary. The prompts used for these examples are reported in Tab. 10 . Relative to representative external baselines and the matched SFT model, our method more consistently preserves subject identity and produces coherent continuations across the boundary.

Dataset	Method	FID	LPIPS full	LPIPS Ω	DINOv2 Ω	DreamSim ctx
Synth-Ads	FLUX.2 SFT	24.07	0.2103	0.3524	0.3134	0.0479
	GPT-Image-2	23.59	0.2083	0.4274	0.3816	0.0454
	Ours	23.88	0.2094	0.3539	0.3154	0.0475
Real-Ads	FLUX.2 SFT	49.25	0.2010	0.3754	0.3200	0.0660
	GPT-Image-2	50.19	0.2135	0.4828	0.4241	0.0677
	Ours	47.31	0.1999	0.3686	0.3156	0.0629

Table 3: Comparison on test samples with available GPT-Image-2 predictions (95.1% overall; 97.2% Synth-Ads, 96.3% Real-Ads, 94.1% LVIS, and 93.6% Open Image). Best in bold ; second best underlined . Lower is better.

Figure 4: Comparison with GPT-Image-2. Representative comparisons illustrating (a) subject-continuation misalignment, and (b) failure to preserve subject structure.

Dataset	Method	Subj. completion (%)	Identity consist. (%)	Context align. (%)	Outpainting realism (%)
Real-Ads	Ours	59.17	58.75	60.21	58.13
Real-Ads	FLUX.2 SFT	40.83	41.25	39.79	41.88
LVIS	Ours	53.73	53.19	54.92	54.00
LVIS	FLUX.2 SFT	46.27	46.81	45.09	46.00

Table 4: Pairwise preference evaluation. Win-and-half-tie rates (%);

50\%

denotes parity. SFT refers to FLUX.2 SFT.

Variant	Overall		Subject Clarity
	FID	LPIPS full	LPIPS Ω	DINOv2 Ω	DreamSim ctx
w/o VLM guidance	50.70	0.2057	0.3865	0.3326	0.0746
w/o mask-weighted flow	48.95	0.2023	0.3689	0.3169	0.0647
w/o subject localization	50.70	0.2032	0.3767	0.3233	0.0665
w/o noise schedule	49.69	0.2012	0.3701	0.3186	0.0657
w/o subband adaptation	48.92	0.2011	0.3742	0.3224	0.0642

Table 5: Component ablation on Real-Ads. Best in bold ; second best underlined . Lower is better.

Figure 5: Representative qualitative ablation. N.S. & S.A. denote noise schedule and subband adaptation, respectively. Additional examples and analysis are provided in Sec. C.5 .

Figure 6: Attention-routing analysis. Direct attention from generated-image queries to the prompt subwords for “foreleg.” SFT spreads this cue broadly into the outpaint region, whereas ours concentrates it on the supported subject continuation. Maps are overlaid on dimmed predictions and span all image queries, whereas the reported metrics use outpaint queries only.

Appendix figures & tables21 assets

Supplementary material from the paper’s appendix.

Appendix

Dataset	Input	Kept	Yield
Synth-Ads	5,659	3,707	65.5%
Real-Ads	1,356	969	71.5%
LVIS	19,808	6,705	33.8%
Open Images	4,650	1,267	27.2%

Appendix

Table 6: High-level filtering statistics. Yield is the percentage of input images retained as finalized pairs.

Dataset	Train	Val.	Test	Total
Synth-Ads	2,595	178	934	3,707
Real-Ads	681	48	240	969
LVIS	4,677	377	1,651	6,705
Open Images	904	65	298	1,267
Total	8,857	668	3,123	12,648

Appendix

Table 7: Realized train/validation/test split sizes after filtering.

Subject area fraction	Crop-fraction range
$[0.05,0.15)$	$[0.10,0.15]$
$[0.15,0.30)$	$[0.15,0.25]$
$[0.30,0.50)$	$[0.25,0.35]$
$[0.50,1.00]$	$[0.35,0.50]$

Appendix

Table 8: Subject-area-dependent crop-fraction sampling policy.

Dataset	$<10\%$	10–20%	20–30%	30–40%	40–50%	$\geq 50\%$
Synth-Ads	0.6	5.7	21.4	37.7	24.8	9.7
Real-Ads	4.9	11.2	18.2	28.4	20.6	16.7
LVIS	6.8	12.2	18.1	24.0	20.2	18.8
Open Images	3.9	8.8	18.5	27.5	24.2	17.0

Appendix

Table 9: Distribution (%) of samples by realized masked fraction of the full image. Rows may differ from 100% by rounding.

Baseline	Prompt(s)
PowerPaint	All rows use the native image-outpainting context token with an empty free-form prompt.
FLUX.1 Fill	(a) dog wearing a black coat, with the dog’s upper body, back, neck, and head continuing from the visible coat. (b) antique handwritten manuscript page and stacked books, with manuscript page with cursive writing and leather-bound books continuing from the visible boundary. (c) seaplane, with the seaplane’s upper fuselage, cockpit windows, wings, tail, and propeller continuing above the floats. (d) giraffe, with the cropped giraffe’s upper neck and head, with surrounding dry savanna brush continuing naturally.
fal/FLUX.2 LoRA	All rows use the checkpoint-specific instruction Fill the green spaces according to the image.
FLUX.2 SFT	(a) Fill the gray region by extending dog wearing a black coat, with the dog’s upper body, back, neck, and head continuing from the visible coat. (b) Fill the gray region by extending antique handwritten manuscript page and stacked books, with manuscript page with cursive writing and leather-bound books continuing from the visible boundary. (c) Fill the gray region by extending seaplane, with the seaplane’s upper fuselage, cockpit windows, wings, tail, and propeller continuing above the floats. (d) Fill the gray region by extending giraffe, with the cropped giraffe’s upper neck and head, with surrounding dry savanna brush continuing naturally.
Ours	Same as FLUX.2 SFT.

Appendix

Table 10: Prompts used for the predictions in the main paper qualitative comparison. Labels (a)–(d) follow the figure rows. Prompt strings are reproduced as used at inference.

Method	Trainable params.	Train complexity	Time/image
FLUX.2 SFT	39.977M	$X$	13.32 s
Ours	39.977M	$X+\mathcal{O}(CHW)$	13.32 s

Appendix

Table 11: Computational cost relative to matched SFT. Complexity denotes per-step training work. Time/image is the sample-weighted wall-clock mean over the test set at 30 denoising steps.

Figure 8: Metric-selection behavior. Each column shows the prediction ranked best by the indicated metric among the methods; the best and second-best methods and scores appear below the image to expose the selection margin. Higher is better for masked PSNR and SSIM, whereas lower is better for LPIPS Ω and DreamSim ctx .

Panel of Figure 10 — Figure 10: Example MLLM judgment on LVIS. Ours is preferred over matched SFT for all four reported criteria. In the anonymous evaluation, FLUX.2 SFT was presented as Method A and ours as Method B; the model identities in parentheses are shown only for the reader. The judge saw the masked input, binary mask, and the two predictions, but not the ground-truth completion. The text below the images condenses the criterion-level rationales returned by GPT-5.5.

Figure 11: Auxiliary high-frequency loss-weight sensitivity. FID for fixed global coefficients and a sample-adaptive coefficient. Lower is better.

Panel of Figure 12 — Figure 12: Two complementary directions of visual attention. Rows compare ours with matched SFT; columns show denoising steps. Both models retrieve evidence from the correct visible subject in (a) , but differ in where they deploy that evidence in (b) .

	Visible-subject routing				“Foreleg” routing
Step	$\mu_{O}$		$R$		$\mu_{O}$		$R$
	SFT	Ours	SFT	Ours	SFT	Ours	SFT	Ours
5	0.0341	0.0277	2.56	3.23	0.00341	0.00056	0.66	0.31
10	0.0251	0.0207	4.03	5.15	0.00238	0.00102	0.67	0.59
20	0.0162	0.0068	4.91	13.54	0.00209	0.00024	0.90	5.39
30	0.0063	0.0026	5.13	11.63	0.00087	0.00022	1.47	4.91

Appendix

Table 12: Step-resolved routing metrics in the final double-stream block. Lower

\mu_{O}

indicates less unsupported-region routing, while higher

R

indicates better localization. Best values within each probe and step are bold.

Panel of Figure 13 — Figure 13: Word-conditioned routing. Direct attention from image queries to selected prompt subwords. The “foreleg” cue is the clearest discriminator: SFT spreads it into unsupported locations, consistent with the duplicated limb, whereas ours increasingly confines it to the valid subject continuation

Panel of Figure 14 — Figure 14: Additional qualitative ablations. N.S. & S.A. denote noise schedule and subband adaptation, respectively

Panel of Figure 15 — Figure 15: Additional qualitative comparisons (Part 1). Columns denote samples and rows denote methods.

Panel of Figure 16 — Figure 16: Additional qualitative comparisons (Part 2). Columns denote samples and rows denote methods.

Panel of Figure 17 — Figure 17: Additional qualitative comparisons (Part 3). Columns denote samples and rows denote methods.

Panel of Figure 18 — Figure 18: Additional qualitative comparisons (Part 4). Columns denote samples and rows denote methods.

Panel of Figure 19 — Figure 19: Additional qualitative comparisons (Part 5). Columns denote samples and rows denote methods.

Panel of Figure 20 — Figure 20: Additional comparisons with GPT-Image-2 (Part 1). Each row is a separate comparison panel; columns show the Input / GT visualization, GPT-Image-2, and our model.

Panel of Figure 21 — Figure 21: Additional comparisons with GPT-Image-2 (Part 2). Each row is a separate comparison panel; columns show the Input / GT visualization, GPT-Image-2, and our model.

Panel of Figure 22 — Figure 22: Additional comparisons with GPT-Image-2 (Part 3). Each row is a separate comparison panel; columns show the Input / GT visualization, GPT-Image-2, and our model.

Image Generation

Image generation studies models that create or edit images from text, examples, layouts, and other structured controls.

See all 474 papers

CopyCat: Improving Fine-Grained Subject Consistency in Subject-to-Image Models within Seconds
Aug 1, 2026Peng Zheng, Ruiqi Liu, Rui Ma +1Diffusion Transformer Subject-Driven Image Generation
School of Artificial Intelligence, Jilin University · 2Shanghai Innovation Institute · 4Engineering Research Center of Knowledge-Driven Human-Machine Intelligence, MOE, China +1
FlowFixer: Towards Detail-Preserving Subject-Driven Generation
Feb 24, 2026Jinyoung Jun, Won-Dong Jang, Wenbin Ouyang +2Subject-Driven Image Generation Image Generation
Amazon · Korea University
High-Resolution Artwork Outpainting with Global Blueprint Guidance and Layout Control
Jul 7, 2026Junha Kim, Hyunjoon Park, Donghyeon ChoDiffusion Model Guidance Image Generation
Hanyang University, Seoul, Republic of Korea

Aug 1, 2026J/K move · Enter open · S save

CopyCat: Improving Fine-Grained Subject Consistency in Subject-to-Image Models within Seconds

Peng Zheng, Ruiqi Liu, Rui Ma, Zuxuan Wu

School of Artificial Intelligence, Jilin University · 2Shanghai Innovation Institute · 4Engineering Research Center of Knowledge-Driven Human-Machine Intelligence, MOE, China+1

Diffusion Transformer Subject-Driven Image Generation Image Generation Diffusion Model Fine-Tuning

Open paper AskPDF arXiv

Recent subject-to-image models have achieved impressive progress in personalized image generation, yet they still struggle to preserve fine-grained subject-specific details. A major reason is the lack of high-quality fine-grained identity supervision: real paired data are expensive to collect, while synthesized training pairs often preserve only coarse subject appearance and fail to capture subtle subject-specific details. In this work, we propose CopyCat, a lightweight model-refinement framework that improves fine-grained subject consistency within only a few seconds. CopyCat performs a one-time refinement of a pretrained subject-to-image model by attaching a lightweight Fine-grained Consistency LoRA (FCLoRA) and optimizing it using a single proxy image, which is used as both the conditioning image and the reconstruction target. This exact self-reconstruction objective substantially simplifies the optimization task, enabling effective fine-grained refinement within only a few seconds. The refinement is performed only once; the resulting model can be directly applied to diverse unseen reference subjects and prompts without further subject-specific optimization. We further revisit subject-to-image LoRA training in double-stream diffusion transformers and find that adapting only the visual stream consistently improves subject consistency. Extensive experiments on DreamBench and XVerseBench demonstrate consistent improvements in fine-grained subject consistency across representative subject-to-image models under both single- and multi-subject settings.

Abstract

Image Generation

Explore similar work

CopyCat: Improving Fine-Grained Subject Consistency in Subject-to-Image Models within Seconds

FlowFixer: Towards Detail-Preserving Subject-Driven Generation

High-Resolution Artwork Outpainting with Global Blueprint Guidance and Layout Control