Video Deblurring with Deconvolution and Aggregation Networks

2025-06-04

AI Summary

Existing video deblurring methods struggle to effectively model inter-frame motion consistency, limiting restoration quality. To address this, we propose the Deconvolution-and-Aggregation Network (DAN), a three-stage collaborative framework comprising preprocessing, alignment-aware deconvolution, and frame aggregation. Specifically, DAN unifies non-local preprocessing, optical-flow-guided learnable frame alignment, multi-scale feature deconvolution, and pixel-wise weighted aggregation guided by a learnable reliability map. This map dynamically estimates the confidence of each neighboring frame’s contribution at every spatial location, enabling adaptive temporal fusion. Extensive experiments on multiple benchmark datasets demonstrate that DAN significantly outperforms state-of-the-art methods, achieving substantial PSNR and SSIM improvements. Qualitative results further reveal superior texture sharpness and enhanced temporal stability in restored videos.

Technology Category

Computer Vision: Video Understanding & Activity AnalysisMachine Learning: Multimodal LearningIntelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphsSearch and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAGResponsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

In contrast to single-image deblurring, video deblurring has the advantage that neighbor frames can be utilized to deblur a target frame. However, existing video deblurring algorithms often fail to properly employ the neighbor frames, resulting in sub-optimal performance. In this paper, we propose a deconvolution and aggregation network (DAN) for video deblurring that utilizes the information of neighbor frames well. In DAN, both deconvolution and aggregation strategies are achieved through three sub-networks: the preprocessing network (PPN) and the alignment-based deconvolution network (ABDN) for the deconvolution scheme; the frame aggregation network (FAN) for the aggregation scheme. In the deconvolution part, blurry inputs are first preprocessed by the PPN with non-local operations. Then, the output frames from the PPN are deblurred by the ABDN based on the frame alignment. In the FAN, these deblurred frames from the deconvolution part are combined into a latent frame according to reliability maps which infer pixel-wise sharpness. The proper combination of three sub-networks can achieve favorable performance on video deblurring by using the neighbor frames suitably. In experiments, the proposed DAN was demonstrated to be superior to existing state-of-the-art methods through both quantitative and qualitative evaluations on the public datasets.

Problem

• Improves video deblurring using neighbor frame information
• Proposes a network combining deconvolution and aggregation strategies
• Enhances performance via alignment-based deconvolution and reliability maps

Innovation

• Uses deconvolution and aggregation networks
• Integrates three specialized sub-networks
• Employs reliability maps for frame combination