Saved in:
Bibliographic Details
Main Authors: Shang, Wei, Ren, Dongwei, Zhang, Wanying, Zhu, Pengfei, Hu, Qinghua, Zuo, Wangmeng
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.07708
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908444575924224
author Shang, Wei
Ren, Dongwei
Zhang, Wanying
Zhu, Pengfei
Hu, Qinghua
Zuo, Wangmeng
author_facet Shang, Wei
Ren, Dongwei
Zhang, Wanying
Zhu, Pengfei
Hu, Qinghua
Zuo, Wangmeng
contents Local motion blur in digital images originates from the relative motion between dynamic objects and static imaging systems during exposure. Existing deblurring methods face significant challenges in addressing this problem due to their inefficient allocation of computational resources and inadequate handling of spatially varying blur patterns. To overcome these limitations, we first propose a trainable mask predictor that identifies blurred regions in the image. During training, we employ blur masks to exclude sharp regions. For inference optimization, we implement structural reparameterization by converting $3\times 3$ convolutions to computationally efficient $1\times 1$ convolutions, enabling pixel-level pruning of sharp areas to reduce computation. Second, we develop an intra-frame motion analyzer that translates relative pixel displacements into motion trajectories, establishing adaptive guidance for region-specific blur restoration. Our method is trained end-to-end using a combination of reconstruction loss, reblur loss, and mask loss guided by annotated blur masks. Extensive experiments demonstrate superior performance over state-of-the-art methods on both local and global blur datasets while reducing FLOPs by 49\% compared to SOTA models (e.g., LMD-ViT). The source code is available at https://github.com/shangwei5/M2AENet.
format Preprint
id arxiv_https___arxiv_org_abs_2507_07708
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Motion-Aware Adaptive Pixel Pruning for Efficient Local Motion Deblurring
Shang, Wei
Ren, Dongwei
Zhang, Wanying
Zhu, Pengfei
Hu, Qinghua
Zuo, Wangmeng
Computer Vision and Pattern Recognition
I.4.3
Local motion blur in digital images originates from the relative motion between dynamic objects and static imaging systems during exposure. Existing deblurring methods face significant challenges in addressing this problem due to their inefficient allocation of computational resources and inadequate handling of spatially varying blur patterns. To overcome these limitations, we first propose a trainable mask predictor that identifies blurred regions in the image. During training, we employ blur masks to exclude sharp regions. For inference optimization, we implement structural reparameterization by converting $3\times 3$ convolutions to computationally efficient $1\times 1$ convolutions, enabling pixel-level pruning of sharp areas to reduce computation. Second, we develop an intra-frame motion analyzer that translates relative pixel displacements into motion trajectories, establishing adaptive guidance for region-specific blur restoration. Our method is trained end-to-end using a combination of reconstruction loss, reblur loss, and mask loss guided by annotated blur masks. Extensive experiments demonstrate superior performance over state-of-the-art methods on both local and global blur datasets while reducing FLOPs by 49\% compared to SOTA models (e.g., LMD-ViT). The source code is available at https://github.com/shangwei5/M2AENet.
title Motion-Aware Adaptive Pixel Pruning for Efficient Local Motion Deblurring
topic Computer Vision and Pattern Recognition
I.4.3
url https://arxiv.org/abs/2507.07708