ODDR: One-Step Deshadow Diffusion via Reward Guidance

Junseong Shin; Kijun Kim; Minseong Kim; Dongjin Kim; Tae Hyun Kim

cs.CVOct 1, 2026

ODDR: One-Step Deshadow Diffusion via Reward Guidance

Authors: Junseong Shin, Kijun Kim, Minseong Kim, Dongjin Kim, Tae Hyun Kim

Organizations: Department of Artificial Intelligence, Hanyang University · AX Future Technology Institute, KT · Department of Computer Science, Hanyang University

Abstract

Recent advances in deep learning for shadow removal have significantly enhanced image quality and realism. However, most approaches rely on real-world paired datasets, which are costly to collect and often limited in scene diversity, leading to limited generalization. To address these limitations, we propose One-step Deshadow Diffusion via Reward guidance (ODDR), a new framework that achieves efficient and high-fidelity shadow removal without relying on real-world paired supervision. Our method begins with One-step Deshadow Diffusion (ODD), a baseline model trained on synthetic shadow data for efficient one-step shadow-free reconstruction. We further adapt ODD into ODDR using ShadowReward. In contrast to traditional, annotation-heavy approaches, ShadowReward is the first reward model for shadow removal trained entirely without human annotation. It learns to mimic human perceptual judgments by ranking synthetically generated images with controlled degradations, such as texture distortion and boundary artifacts. This reward-guided fine-tuning enables ODDR to close the synthetic-to-real domain gap. Extensive experiments show that ODD achieves strong performance without relying on real-world paired supervision, and ODDR further improves the results, narrowing the gap to fully supervised methods trained on real-world paired data while maintaining higher computational efficiency as a single-step model.

Figures & tables

Figure 1: We compare our method with diffusion-based shadow removal models, DeS3 [ 22 ] and SSR [ 62 ] on SRD [ 44 ] , LRSS [ 9 ] , and UIUC [ 14 ] datasets, and analyze computational efficiency. Our ODDR achieves the highest restoration quality and simultaneously demands the least FLOPs, inference time, and NFE (Number of Function Evaluations).

Figure 2: Overall training pipeline. (a) The baseline ODD is trained on synthetic shadow images ( $x_{\text{syn}}$ ), optimizing the ODD-LoRA and ODD-Conv modules. (b) ODDR is then fine-tuned from ODD parameters using real-world shadow images ( $x_{\text{real}}$ ) guided by ShadowReward ( $r_{\theta}$ ) model, optimizing ODDR-LoRA module.

Figure 3: An example list $\{x^{(j)}\}_{j=0}^{3}$ of the human-annotation-free dataset. Larger $j$ means more degradation.

$\mathcal{D}$	Methods	Mask-free	Shadow Region			Non-Shadow Region			All Image
$\mathcal{D}$	Methods	Mask-free	PSNR $\uparrow$	SSIM $\uparrow$	RMSE $\downarrow$	PSNR $\uparrow$	SSIM $\uparrow$	RMSE $\downarrow$	PSNR $\uparrow$	SSIM $\uparrow$	RMSE $\downarrow$
	Real-world Shadow Image	–	20.83	0.927	39.01	37.46	0.985	2.40	20.46	0.894	8.40
w/ real-world pair	DHAN [ 7 ]	✗	32.92	0.988	9.6	27.15	0.971	7.4	25.66	0.956	7.8
	SP+M-Net [ 27 ]	✗	37.60	0.990	6.3	36.02	0.976	3.0	32.94	0.962	3.5
	Fu et al. [ 8 ]	✗	36.04	0.978	6.7	31.16	0.892	3.8	29.45	0.861	4.2
	BMNet [ 68 ]	✗	37.87	0.991	5.62	37.51	0.985	2.45	33.98	0.972	2.97

Table 1: Quantitative comparison on AISTD [ 27 ] dataset. Red and blue denote the best and second-best results, evaluated among the methods trained without real-world pairs. ✓ denotes mask-free methods and ✗ indicates methods that require shadow masks.

\mathcal{D}

indicates whether real-world paired data is used for training.

Figure 5: Visual comparison on AISTD [ 27 ] . Top: non-shadow preservation (yellow arrows mark competitor failures). Bottom: shadow removal. Ours performs well in both.

	SRD [ 44 ]			LRSS [ 9 ]			UIUC [ 14 ]
Method	PSNR $\uparrow$	SSIM $\uparrow$	RMSE $\downarrow$	PSNR $\uparrow$	SSIM $\uparrow$	RMSE $\downarrow$	PSNR $\uparrow$	SSIM $\uparrow$	RMSE $\downarrow$
S3R-Net [ 25 ]	22.51	0.878	10.60	22.54	0.786	10.45	23.92	0.857	8.53
DCS [ 21 ]	21.99	0.868	11.47	22.10	0.774	11.63	23.94	0.852	9.69
G2R [ 36 ]	24.23	0.898	8.40	22.79	0.778	9.64	27.03	0.867	6.23
DeS3 [ 22 ]	22.44	0.900	10.29	21.10	0.767	12.75	19.02	0.822	11.23
SSR [ 62 ]	25.86	0.922	8.25	24.47	0.790	9.89	27.79	0.873	7.61

Table 2: Results on SRD, LRSS, and UIUC. The best results are highlighted in red .

Figure 6: Visual comparisons on SRD [ 44 ] , LRSS [ 9 ] , and UIUC [ 14 ] (top to bottom).

Table 9

Table 10

Appendix figures & tables17 assets

Supplementary material from the paper’s appendix.

Appendix

Type	Operation	$j=1$	$j=2$	$j=3$
T	Gaussian blur	$3{\times}3$	$5{\times}5$	$7{\times}7$
N	Gaussian noise	$\sigma{=}0.004$	$\sigma{=}0.008$	$\sigma{=}0.012$
C	Hue shift	$\Delta h{=}\pm 0.01$	$\Delta h{=}\pm 0.02$	$\Delta h{=}\pm 0.03$
	Saturation scale	$s{\in}[0.95,1.05]$	$s{\in}[0.90,1.10]$	$s{\in}[0.85,1.15]$
	Brightness scale	$b{\in}[0.97,1.03]$	$b{\in}[0.94,1.06]$	$b{\in}[0.92,1.08]$
L	Gamma correction	$\gamma{\in}[0.97,1.03]$	$\gamma{\in}[0.94,1.06]$	$\gamma{\in}[0.90,1.10]$

Appendix

Table 7: Parameter ranges used to synthesize cumulative degradations (T, N, C, L, B) for the Human-Annotation-Free Dataset. Higher

j

corresponds to stronger degradation (

j=0

is the clean image). Here,

r

denotes the boundary dilation radius,

\Delta h

the hue shift,

s

the saturation scaling factor,

b

the brightness scaling factor, and

\gamma

the luminance gamma parameter. All parameters are uniformly sampled within the specified ranges. Larger values (or wider ranges) imply more severe distortions.

List length ( $J$ )	Shadow		All
List length ( $J$ )	PSNR $\uparrow$	RMSE $\downarrow$	PSNR $\uparrow$	RMSE $\downarrow$
1	37.29	5.99	33.70	3.06
2	37.78	5.74	33.96	3.01
3	38.37	5.36	34.20	2.97
4	38.17	5.45	34.15	2.97

Appendix

Table 8: Impact of the degradation list length

J

on shadow removal performance.

Figure 7: User study interface for human preference acquisition. Annotators rank three randomized results (A, B, C) against the input image (red border).

Figure 8: Comparison of reward model ranking accuracy. ShadowReward correctly ranks the shadow removal results in alignment with human preference ( $A>B>C$ ), while existing reward models fail to match human perception.

Figure 9: PSNR is unreliable under GT misalignment, while our ShadowReward (labeled ‘Reward’) provides perceptually consistent evaluation.

Backbone	Shadow		All
Backbone	PSNR $\uparrow$	RMSE $\downarrow$	PSNR $\uparrow$	RMSE $\downarrow$
CLIP	37.27	5.91	33.67	3.04
MoCov3	37.72	5.71	33.91	3.01
DINOv2	38.37	5.36	34.20	2.97

Appendix

Table 9: Impact of different feature extractor backbones for ShadowReward.

Method	Shadow		All
Method	PSNR $\uparrow$	RMSE $\downarrow$	PSNR $\uparrow$	RMSE $\downarrow$
Mean	37.66	5.77	33.90	3.02
Max	37.62	5.56	33.89	3.02
Top- $K$ (0.05)	38.28	5.48	34.19	2.97
Top- $K$ (0.15)	38.08	5.51	34.12	2.98
Top- $K$ (0.1)	38.37	5.36	34.20	2.97

Appendix

Table 10: Impact of different spatial reduction methods for ShadowReward.

Guidance Method	Shadow		ALL
Guidance Method	PSNR $\uparrow$	RMSE $\downarrow$	PSNR $\uparrow$	RMSE $\downarrow$
Shadow Detection	36.69	6.26	33.33	3.11
Conditional GAN	36.79	6.40	33.36	3.12
DINO Discriminator	37.79	5.74	33.91	3.03
ShadowReward (ours)	38.37	5.36	34.20	2.97

Appendix

Table 11: Comparison with alternative guidance strategies on the AISTD dataset. All methods use the same ODD baseline and unpaired fine-tuning protocol.

Method	ODD Train	Shadow		ALL
Method	ODD Train	PSNR $\uparrow$	RMSE $\downarrow$	PSNR $\uparrow$	RMSE $\downarrow$
ODD	AISTD	36.67	6.27	33.32	3.11
ODDR	AISTD	38.37	5.36	34.20	2.97
ODD	USR	36.08	6.70	32.86	3.24
ODDR	USR	37.58	5.80	33.71	3.13

Appendix

Table 12: Effect of training data source overlap on the AISTD dataset. USR denotes a disjoint shadow-free dataset used to train ODD.

Figure 10: Visual comparisons of our ODD and ODDR on the AISTD [ 27 ] (top), LRSS [ 9 ] (middle), and UIUC [ 14 ] (bottom) datasets.

Figure 11: Detailed illustration of VAE skip-connection.

Figure 12: Example of failure case on highly textured/non-flat surfaces.

Figure 13: Additional visual comparison on the AISTD [ 27 ] dataset.

Figure 14: Additional visual comparisons on the LRSS [ 9 ] (top four rows) and UIUC [ 14 ] (bottom four rows) datasets.

Figure 15: Additional visual comparison of ODD and ODDR on the AISTD [ 27 ] dataset.

Figure 16: Additional visual comparison of ODD and ODDR on the unseen LRSS [ 9 ] dataset.

Figure 17: Additional visual comparison of ODD and ODDR on the unseen UIUC [ 14 ] dataset.

Explore similar work

Shadow Feature Refinement Network: Progressive Feature Refinement based on Knowledge Distillation for Effective Shadow Removal
Oct 4, 2026Donghyun Han, Byoung-Dai LeeVideo Object Removal
Kyonggi University, Korea
FreeShadow: Training-Free Shadow Removal via Illumination Transfer and Selective Content Preservation in Diffusion Models
Jul 29, 2026Yinan Wang, Yan Huang, Yong Xu +1Diffusion Models Illumination Variations
School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China · Pazhou Lab, Guangzhou 510005, China · Polytech Nantes, Universit´e de Nantes, Nantes 44306, France
Winner of CVPR2026 NTIRE Challenge on Image Shadow Removal: Semantic and Geometric Guidance for Shadow Removal via Cascaded Refinement
Apr 17, 2026Lorenzo Beltrame, Jules Salzinger, Filip Svoboda +4Video Object Removal Cvpr
1Austrian Institute of Technology · 2Technical University of Munich · University of Cambridge +1

Oct 4, 2026, cs.CVJ/K move · Enter open · S save

Shadow Feature Refinement Network: Progressive Feature Refinement based on Knowledge Distillation for Effective Shadow Removal

Donghyun Han, Byoung-Dai Lee

Kyonggi University, Korea

Video Object Removal

Open paper AskPDF arXiv

In the field of deep learning, has seen significant advancements; however, shadow removal remains a persistent challenge owing to the variable sizes and colors of shadows influenced by lighting conditions. This study proposes a novel shadow feature refinement network (SFR-Net), which leverages supervised learning, feature refinement loss, and knowledge distillation to enhance shadow removal performance. A dedicated post-processing algorithm is further introduced to restore natural color consistency in the generated shadow-free images. We evaluated our method on two public datasets: the adjusted image shadow triplet dataset (ISTD+) and the shadow removal dataset (SRD), which demonstrate strong generalization capabilities under diverse conditions. On ISTD+, our model achieved a root mean square error (RMSE) of 3.4627 and structural similarity index measure (SSIM) of 0.9382 across the entire image. On SRD, it recorded an RMSE of 4.3781 and an SSIM of 0.9341. These comprehensive results show that our approach performs competitively across both shadow and non-shadow regions while setting a promising direction for robust and perceptually natural shadow removal. Code is available at https://github.com/DongHyun99/SFRNet.

Abstract

Explore similar work

Shadow Feature Refinement Network: Progressive Feature Refinement based on Knowledge Distillation for Effective Shadow Removal

FreeShadow: Training-Free Shadow Removal via Illumination Transfer and Selective Content Preservation in Diffusion Models

Winner of CVPR2026 NTIRE Challenge on Image Shadow Removal: Semantic and Geometric Guidance for Shadow Removal via Cascaded Refinement