Unraveling Motion Uncertainty for Local Motion Deblurring

Zeyu Xiao, Zhihe Lu, Michael Bi Mi, Zhiwei Xiong, Xinchao Wang

Abstract: In real-world photography, local motion blur often arises from the interplay between moving objects and stationary backgrounds during exposure. Existing deblurring methods face challenges in addressing local motion deblurring due to (i) the presence of arbitrary localized blurs and uncertain blur extents; (ii) the limited ability to accurately identify specific blurs resulting from ambiguous motion boundaries. These limitations often lead to suboptimal solutions when estimating blur maps and generating final deblurred images. To that end, we propose a novel method named Motion-Uncertainty-Guided Network (MUGNet), which harnesses a probabilistic representational model to explicitly address the intricacies stemming from motion uncertainties. Specifically, MUGNet consists of two key components, i.e., motion-uncertainty quantification (MUQ) module and motion-masked separable attention (M2SA) module, serving for complementary purposes. Concretely, MUQ aims to learn a conditional distribution for accurate and reliable blur map estimation, while the M2SA module is to enhance the representation of regions influenced by local motion blur and static background, which is achieved by promoting the establishment of extensive global interactions. We demonstrate the superiority of our MUGNet with extensive experiments.

Primary Subject Area: [Experience] Multimedia Applications

Secondary Subject Area: [Content] Media Interpretation

Relevance To Conference: This work significantly contributes to multimedia and multimodal processing by addressing the challenge of local motion deblurring in real-world photography. Through careful designs and extensive experiments, we demonstrate the superiority of the proposed method in effectively addressing local motion deblurring challenges. This advancement in local motion deblurring has wide-ranging implications for multimedia applications, improving the visual quality and interpretability of images in various real-world scenarios.

Submission Number: 365