BaMBNet: A Blur-Aware Multi-Branch Network for Dual-Pixel Defocus Deblurring

doi:10.1109/JAS.2022.105563

CASIA OpenIR > 学术期刊 > IEEE/CAA Journal of Automatica Sinica

	BaMBNet: A Blur-Aware Multi-Branch Network for Dual-Pixel Defocus Deblurring
	Pengwei Liang; Junjun Jiang; Xianming Liu; Jiayi Ma
发表期刊	IEEE/CAA Journal of Automatica Sinica
ISSN	2329-9266
	2022
卷号	9 期号:5 页码:878-892
摘要	Reducing the defocus blur that arises from the finite aperture size and short exposure time is an essential problem in computational photography. It is very challenging because the blur kernel is spatially varying and difficult to estimate by traditional methods. Due to its great breakthrough in low-level tasks, convolutional neural networks (CNNs) have been introduced to the defocus deblurring problem and achieved significant progress. However, previous methods apply the same learned kernel for different regions of the defocus blurred images, thus it is difficult to handle nonuniform blurred images. To this end, this study designs a novel blur-aware multi-branch network (BaMBNet), in which different regions are treated differentially. In particular, we estimate the blur amounts of different regions by the internal geometric constraint of the dual-pixel (DP) data, which measures the defocus disparity between the left and right views. Based on the assumption that different image regions with different blur amounts have different deblurring difficulties, we leverage different networks with different capacities to treat different image regions. Moreover, we introduce a meta-learning defocus mask generation algorithm to assign each pixel to a proper branch. In this way, we can expect to maintain the information of the clear regions well while recovering the missing details of the blurred regions. Both quantitative and qualitative experiments demonstrate that our BaMBNet outperforms the state-of-the-art (SOTA) methods. For the dual-pixel defocus deblurring (DPD)-blur dataset, the proposed BaMBNet achieves 1.20 dB gain over the previous SOTA method in term of peak signal-to-noise ratio (PSNR) and reduces learnable parameters by 85%. The details of the code and dataset are available at https://github.com/junjun-jiang/BaMBNet.
关键词	Blur kernel convolutional neural networks (CNNs) defocus deblurring dual-pixel (DP) data meta-learning
DOI	10.1109/JAS.2022.105563
引用统计	被引频次：7[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/47551
专题	学术期刊_IEEE/CAA Journal of Automatica Sinica
推荐引用方式 GB/T 7714	Pengwei Liang,Junjun Jiang,Xianming Liu,et al. BaMBNet: A Blur-Aware Multi-Branch Network for Dual-Pixel Defocus Deblurring[J]. IEEE/CAA Journal of Automatica Sinica,2022,9(5):878-892.
APA	Pengwei Liang,Junjun Jiang,Xianming Liu,&Jiayi Ma.(2022).BaMBNet: A Blur-Aware Multi-Branch Network for Dual-Pixel Defocus Deblurring.IEEE/CAA Journal of Automatica Sinica,9(5),878-892.
MLA	Pengwei Liang,et al."BaMBNet: A Blur-Aware Multi-Branch Network for Dual-Pixel Defocus Deblurring".IEEE/CAA Journal of Automatica Sinica 9.5(2022):878-892.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
JAS-2022-0100.pdf（12217KB）	期刊论文	出版稿	开放获取	CC BY-NC-SA	浏览