Saved in:
Bibliographic Details
Main Authors: Liu, Yike, Zhang, Jianhui, Li, Haipeng, Liu, Shuaicheng, Zeng, Bing
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.12222
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909581846773760
author Liu, Yike
Zhang, Jianhui
Li, Haipeng
Liu, Shuaicheng
Zeng, Bing
author_facet Liu, Yike
Zhang, Jianhui
Li, Haipeng
Liu, Shuaicheng
Zeng, Bing
contents While recent video deblurring methods have advanced significantly, they often overlook two valuable prior information: (1) motion vectors (MVs) and coding residuals (CRs) from video codecs, which provide efficient inter-frame alignment cues, and (2) the rich real-world knowledge embedded in pre-trained diffusion generative models. We present CPGDNet, a novel two-stage framework that effectively leverages both coding priors and generative diffusion priors for high-quality deblurring. First, our coding-prior feature propagation (CPFP) module utilizes MVs for efficient frame alignment and CRs to generate attention masks, addressing motion inaccuracies and texture variations. Second, a coding-prior controlled generation (CPC) module network integrates coding priors into a pretrained diffusion model, guiding it to enhance critical regions and synthesize realistic details. Experiments demonstrate our method achieves state-of-the-art perceptual quality with up to 30% improvement in IQA metrics. Both the code and the codingprior-augmented dataset will be open-sourced.
format Preprint
id arxiv_https___arxiv_org_abs_2504_12222
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Coding-Prior Guided Diffusion Network for Video Deblurring
Liu, Yike
Zhang, Jianhui
Li, Haipeng
Liu, Shuaicheng
Zeng, Bing
Computer Vision and Pattern Recognition
While recent video deblurring methods have advanced significantly, they often overlook two valuable prior information: (1) motion vectors (MVs) and coding residuals (CRs) from video codecs, which provide efficient inter-frame alignment cues, and (2) the rich real-world knowledge embedded in pre-trained diffusion generative models. We present CPGDNet, a novel two-stage framework that effectively leverages both coding priors and generative diffusion priors for high-quality deblurring. First, our coding-prior feature propagation (CPFP) module utilizes MVs for efficient frame alignment and CRs to generate attention masks, addressing motion inaccuracies and texture variations. Second, a coding-prior controlled generation (CPC) module network integrates coding priors into a pretrained diffusion model, guiding it to enhance critical regions and synthesize realistic details. Experiments demonstrate our method achieves state-of-the-art perceptual quality with up to 30% improvement in IQA metrics. Both the code and the codingprior-augmented dataset will be open-sourced.
title Coding-Prior Guided Diffusion Network for Video Deblurring
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.12222