V-Bridge: Bridging Video Generative Priors to Versatile Few-shot Image Restoration

Shenghe Zheng; Junpeng Jiang; Wenbo Li

cs.CVMar 13, 2026

V-Bridge: Bridging Video Generative Priors to Versatile Few-shot Image Restoration

Authors: Shenghe Zheng, Junpeng Jiang, Wenbo Li

Organizations: The Hong Kong University of Science and Technology · Harbin Institute of Technology, Shenzhen · The Chinese University of Hong Kong

Abstract

Large-scale video generative models are trained on vast and diverse visual data, enabling them to internalize rich structural, semantic, and dynamic priors of the visual world. While these models have demonstrated impressive generative capability, their potential as general-purpose visual learners remains largely untapped. In this work, we introduce V-Bridge, a framework that bridges this latent capacity to versatile few-shot image restoration tasks. We reinterpret image restoration not as a static regression problem, but as a progressive generative process, and leverage video models to simulate the gradual refinement from degraded inputs to high-fidelity outputs. Surprisingly, with only 1,000 multi-task training samples (less than 2% of existing restoration methods), pretrained video models can be induced to perform competitive image restoration, achieving multiple tasks with a single model, rivaling specialized architectures designed explicitly for this purpose. Our findings reveal that video generative models implicitly learn powerful and transferable restoration priors that can be activated with only extremely limited data, challenging the traditional boundary between generative modeling and low-level vision, and opening a new design paradigm for foundation models in visual tasks.

Figures & tables

Figure 1: Left: Image restoration is formulated as progressive video generation with frame drift correction. Right: Leveraging video generative priors leads to stronger generalization under limited data compared to current image restoration method [ 25 ] .

Figure 2: Overview of the proposed pipeline. The upper part shows data construction and training, where paired low- and high-quality images are used to build pseudo-temporal sequences for progressive restoration learning. A progressive resolution training strategy is adopted to improve fine-grained detail modeling, and an auxiliary generative model is trained for final-frame correction. The lower part shows inference, where the model generates a restoration trajectory and uses the refined last frame as the final output.

Figure 3: Visualization results on a subset of the FoundIR test set. FoundIR-G is the generalist model of FoundIR. GT denotes ground truth. Bounding boxes with different colors indicate zoomed regions for detailed comparison. Compared with other methods, our approach achieves higher visual fidelity and stronger structural consistency, while showing superior robustness across diverse degradation patterns.

Methods		PromptIR	TransWeather	DA-CLIP	DiffUIR	AutoDIR	InstructIR	X-Restormer	AgenticIR	FoundIR-G	Ours
Data Scale $\downarrow$		77K	19K	53K	30K	23K	80K	40K	15K	1M	1K
Dense-Haze	PSNR $\uparrow$	9.57	10.51	10.94	9.59	12.33	11.03	9.57	10.11	9.29	11.97
Dense-Haze	SSIM $\uparrow$	0.4333	0.4523	0.459	0.4326	0.4862	0.4649	0.4295	0.3884	0.4307	0.4897
UHD-LL	PSNR $\uparrow$	11.76	12.3	18.51	11.27	22.52	20.03	11.56	12.82	10.61	17.87
UHD-LL	SSIM $\uparrow$	0.6170	0.6557	0.8120	0.5975	0.8572	0.7356	0.6113	0.6649	0.5775	0.7999
NH-Haze	PSNR $\uparrow$	11.38	11.58	12.35	11.39	12.71	12.24	11.36	12.2	11.43	12.97

Table 2: Quantitative comparisons (

\frac{\mathrm{PSNR}}{\mathrm{SSIM}}

) on the public benchmarks. Underlining and boldface indicate the second-best and best methods, respectively.

Figure 4: Performance improvement on the test dataset of FoundIR brought by few-shot training.

Figure 5: (a) Comparison on the FoundIR test set with and without the refine model. GT denotes ground truth. Correction improves visual quality and enhances fine details. (b) Ablation study results. Top: effect of different training frame numbers for image restoration. Bottom: performance improves as training data scale increases.

Methods		PromptIR	TransWeather	DA-CLIP	DiffUIR	X-Restormer	AgenticIR	FoundIR-G	Ours
Data Scale $\downarrow$		77K	19K	53K	30K	40K	15K	1M	1K
WeatherBench	PSNR $\uparrow$	21.54	20.94	21.59	21.68	21.57	20.35	21.57	20.88
WeatherBench	SSIM $\uparrow$	0.7767	0.7579	0.7731	0.7795	0.7813	0.7304	0.7794	0.7105

Table 6: Quantitative comparisons (

\frac{\mathrm{PSNR}}{\mathrm{SSIM}}

) on the unseen task for our method about desnowing. Underlining and boldface indicate the second-best and best methods, respectively. FoundIR-G denotes the generalist version of FoundIR.

Figure 6: Visualization results on an OOD snow removal scenario. FoundIR-G denotes the Generalist model of FoundIR. GT denotes ground truth. Both methods are evaluated under OOD conditions. Although our model is trained with only 1K samples, it shows significantly better generalization than FoundIR-G trained with 1M images.

Appendix figures & tables5 assets

Supplementary material from the paper’s appendix.

Appendix

Parameter	Value
BF16	True
Learning Rate	2e-5
LR Scheduler Type	constant_with_warmup
LR Warmup Steps	100
Adam Weight Decay	3e-2
Adam Epsilon	1e-10

Appendix

Table 7: Training Configurations for V-Bridge.

Figure 7: Additional visualization results.

Figure 8: Additional visualization results.

Figure 9: Additional visualization results.

Image Restoration

Image restoration uses computational methods to recover useful visual content from noisy, blurred, low-resolution, or otherwise degraded images.

See all 372 papers

LPM: Industrial-Scale Generative Video Restoration
Jul 15, 2026Bichuan Zhu, Fulin Li, Jiachao Gong +14Video Diffusion Models Video Restoration
LPM Team, Kuaishou Technology
Taming Noise Initialization for Zero-shot Video Restoration with Image Diffusion Priors
Oct 27, 2026Chong Wang, Zixuan Fu, Hao Cheng +3Video Restoration Diffusion Models
School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore · School of Artificial Intelligence, Hebei University of Technology, Tianjin, China · Division of Robotics, Perception and Learning, KTH Royal Institute of Technology, Stockholm, Sweden +1
Zero-Shot Video Restoration and Enhancement with Text-to-Image Latent Diffusion Models and Multi-Modal References
Aug 27, 2026Cong Cao, Huanjing Yue, Xin Liu +1Video Restoration T2I Diffusion Models
School of Electrical and Information Engineering, Tianjin University, Tianjin, China · Computer Vision and Pattern Recognition Laboratory, School of Engineering Science, Lappeenranta-Lahti University of Technology LUT, Lappeenranta, Finland

Jul 15, 2026J/K move · Enter open · S save

LPM: Industrial-Scale Generative Video Restoration

Bichuan Zhu, Fulin Li, Jiachao Gong, Jinhua Hao+13

LPM Team, Kuaishou Technology

Video Diffusion Models Video Restoration

Open paper AskPDF arXiv

We present the Large Processing Model (LPM), a diffusion-based generative framework for photorealistic video restoration under complex, in-the-wild degradations. To our knowledge, LPM is the first generative video restoration model deployed at industrial scale. LPM addresses the diverse degradations in user-generated content (UGC) through a unified system encompassing large-scale data engineering, foundation-model training, and efficient inference. Its enhanced architecture, progressive training strategy, and temporal-pyramid inference mechanism jointly enable high-fidelity, temporally consistent restoration of arbitrarily long videos across the broad content distribution encountered on UGC platforms. LPM has been deployed in production at Kuaishou, where videos processed by the model account for approximately 45% of total viewing time, delivering consistent improvements across key quality-of-experience metrics. Beyond perceptual enhancement, LPM delivers substantial system-level benefits: at comparable perceptual quality, it reduces bitrate by 20% relative to Kuaishou's in-house codec, yielding annual bandwidth cost savings on the order of hundreds of millions. Its low serving cost also enables integration into products such as Kling, demonstrating that generative restoration can be practical, scalable, and cost-effective for large-scale video processing.

Abstract

Image Restoration

Explore similar work

LPM: Industrial-Scale Generative Video Restoration

Taming Noise Initialization for Zero-shot Video Restoration with Image Diffusion Priors

Zero-Shot Video Restoration and Enhancement with Text-to-Image Latent Diffusion Models and Multi-Modal References