Filters








36 Hits in 1.2 sec

Extreme Channel Prior Embedded Network for Dynamic Scene Deblurring [article]

Jianrui Cai, Wangmeng Zuo, Lei Zhang
2019 arXiv   pre-print
Recent years have witnessed the significant progress on convolutional neural networks (CNNs) in dynamic scene deblurring. While CNN models are generally learned by the reconstruction loss defined on training data, incorporating suitable image priors as well as regularization terms into the network architecture could boost the deblurring performance. In this work, we propose an Extreme Channel Prior embedded Network (ECPeNet) to plug the extreme channel priors (i.e., priors on dark and bright
more » ... nnels) into a network architecture for effective dynamic scene deblurring. A novel trainable extreme channel prior embedded layer (ECPeL) is developed to aggregate both extreme channel and blurry image representations, and sparse regularization is introduced to regularize the ECPeNet model learning. Furthermore, we present an effective multi-scale network architecture that works in both coarse-to-fine and fine-to-coarse manners for better exploiting information flow across scales. Experimental results on GoPro and Kohler datasets show that our proposed ECPeNet performs favorably against state-of-the-art deep image deblurring methods in terms of both quantitative metrics and visual quality.
arXiv:1903.00763v1 fatcat:vtko6lhyi5cancefrjhxocflz4

Toward Real-World Single Image Super-Resolution: A New Benchmark and A New Model [article]

Jianrui Cai, Hui Zeng, Hongwei Yong, Zisheng Cao, Lei Zhang
2019 arXiv   pre-print
Most of the existing learning-based single image superresolution (SISR) methods are trained and evaluated on simulated datasets, where the low-resolution (LR) images are generated by applying a simple and uniform degradation (i.e., bicubic downsampling) to their high-resolution (HR) counterparts. However, the degradations in real-world LR images are far more complicated. As a consequence, the SISR models trained on simulated data become less effective when applied to practical scenarios. In
more » ... paper, we build a real-world super-resolution (RealSR) dataset where paired LR-HR images on the same scene are captured by adjusting the focal length of a digital camera. An image registration algorithm is developed to progressively align the image pairs at different resolutions. Considering that the degradation kernels are naturally non-uniform in our dataset, we present a Laplacian pyramid based kernel prediction network (LP-KPN), which efficiently learns per-pixel kernels to recover the HR image. Our extensive experiments demonstrate that SISR models trained on our RealSR dataset deliver better visual quality with sharper edges and finer textures on real-world scenes than those trained on simulated datasets. Though our RealSR dataset is built by using only two cameras (Canon 5D3 and Nikon D810), the trained model generalizes well to other camera devices such as Sony a7II and mobile phones.
arXiv:1904.00523v1 fatcat:cy2pud6ghnc37bjz22d2ha7p4m

Learning a Single Tucker Decomposition Network for Lossy Image Compression with Multiple Bits-Per-Pixel Rates [article]

Jianrui Cai, Zisheng Cao, Lei Zhang
2018 arXiv   pre-print
Lossy image compression (LIC), which aims to utilize inexact approximations to represent an image more compactly, is a classical problem in image processing. Recently, deep convolutional neural networks (CNNs) have achieved interesting results in LIC by learning an encoder-quantizer-decoder network from a large amount of data. However, existing CNN-based LIC methods usually can only train a network for a specific bits-per-pixel (bpp). Such a "one network per bpp" problem limits the generality
more » ... d flexibility of CNNs to practical LIC applications. In this paper, we propose to learn a single CNN which can perform LIC at multiple bpp rates. A simple yet effective Tucker Decomposition Network (TDNet) is developed, where there is a novel tucker decomposition layer (TDL) to decompose a latent image representation into a set of projection matrices and a core tensor. By changing the rank of the core tensor and its quantization, we can easily adjust the bpp rate of latent image representation within a single CNN. Furthermore, an iterative non-uniform quantization scheme is presented to optimize the quantizer, and a coarse-to-fine training strategy is introduced to reconstruct the decompressed images. Extensive experiments demonstrate the state-of-the-art compression performance of TDNet in terms of both PSNR and MS-SSIM indices.
arXiv:1807.03470v1 fatcat:ozqw3wnytfazdhvcl6flgwlrx4

CameraNet: A Two-Stage Framework for Effective Camera ISP Learning [article]

Zhetong Liang, Jianrui Cai, Zisheng Cao, Lei Zhang
2019 arXiv   pre-print
This drawback impedes the reconstruction quality for challenging scenarios, resulting in Zhetong Liang, Jianrui Cai images with high noise level, low dynamic range and less vivid color.  ... 
arXiv:1908.01481v2 fatcat:yvmg66m3pjfvjmmtz26qdkaxse

Autophagy lessens ischemic liver injury by reducing oxidative damage

Kai Sun, Xuqin Xie, Yan Liu, Zhipeng Han, Xue Zhao, Ning Cai, Shanshan Zhang, Jianrui Song, Lixin Wei
2013 Cell & Bioscience  
Hepatic ischemia/reperfusion is a multi-factorial process which causes liver injury. It is reported that ischemia alone is sufficient to induce liver injury. Nutrient deprivation is a crucial factor impacting ischemic injury of the liver. Therefore, we explored the role of autophagy in ischemia through using hepatic ischemia rat model in vivo and nutrient-free model in vitro. Results: We found that both ischemia in vivo and nutrient deprivation in vitro activated autophagy, inhibition of which
more » ... ggravated ischemia-or nutrient deficiency-induced injury. In the nutrient-free condition, autophagy inhibition enhanced liver cell necrosis but not apoptosis by promoting reactive oxygen species (ROS) accumulation, and antioxidant NAC could reverse this trend. Inhibition of autophagy also resulted in the increase of the percentage of necrotic cell but not apoptotic cell in the ischemia-treated rat livers. Further studies showed that under nutrient deprivation, autophagy inhibition promoted mitochondrial ROS generation, which further aggravated mitochondria damage. These changes formed a "vicious cycle" that accelerated the process of cell necrosis. Autophagy inhibition also increased mitochondrial oxidative stress during hepatic ischemia, and antioxidant could suppress the aggravation of ischemia-induced liver damage in the co-treatment of autophagy inhibitor. Conclusions: Taken together, our results suggested that autophagy suppressed ischemic liver injury by reducing ROS-induced necrosis. This finding will contribute to the development of the therapeutic strategy about the pre-treatment of liver surgery.
doi:10.1186/2045-3701-3-26 pmid:23758862 pmcid:PMC3693965 fatcat:d4w3yoymlrgbzl55l2wbmo7ypq

Second-Order Attention Network for Single Image Super-Resolution

Tao Dai, Jianrui Cai, Yongbing Zhang, Shu-Tao Xia, Lei Zhang
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Recently, deep convolutional neural networks (CNNs) have been widely explored in single image super-resolution (SISR) and obtained remarkable performance. However, most of the existing CNN-based SISR methods mainly focus on wider or deeper architecture design, neglecting to explore the feature correlations of intermediate layers, hence hindering the representational power of CNNs. To address this issue, in this paper, we propose a second-order attention network (SAN) for more powerful feature
more » ... pression and feature correlation learning. Specifically, a novel trainable second-order channel attention (SOCA) module is developed to adaptively rescale the channel-wise features by using second-order feature statistics for more discriminative representations. Furthermore, we present a non-locally enhanced residual group (NLRG) structure, which not only incorporates non-local operations to capture long-distance spatial contextual information, but also contains repeated local-source residual attention groups (LSRAG) to learn increasingly abstract feature representations. Experimental results demonstrate the superiority of our SAN network over state-of-the-art SISR methods in terms of both quantitative metrics and visual quality. * The first two authors contribute equally to this work.
doi:10.1109/cvpr.2019.01132 dblp:conf/cvpr/DaiCZXZ19 fatcat:tylc6qiuhrbqdju4w5os5vvika

Paradoxical roles of autophagy in different stages of tumorigenesis: protector for normal or cancer cells

Kai Sun, Weijie Deng, Shanshan Zhang, Ning Cai, Shufan Jiao, Jianrui Song, Lixin Wei
2013 Cell & Bioscience  
Autophagy serves as a dynamic degradation and recycling system that provides biological materials and energy in response to stress. The role of autophagy in tumor development is complex. Various studies suggest that autophagy mainly contributes to tumor suppression during the early stage of tumorigenesis and tumor promotion during the late stage of tumorigenesis. During the tumorization of normal cells, autophagy protects genomic stability by retarding stem cells-involved damage/repair cycle,
more » ... d inhibits the formation of chronic inflammatory microenvironment, thus protecting normal cell homeostasis and preventing tumor generation. On the other hand, autophagy also protects tumor cells survival during malignant progression by supporting cellular metabolic demands, decreasing metabolic damage and supporting anoikis resistance and dormancy. Taken together, autophagy appears to play a role as a protector for either normal or tumor cells during the early or late stage of tumorigenesis, respectively. The process of tumorigenesis perhaps needs to undergo twice autophagy-associated screening. The normal cells that have lower autophagy capacity are prone to tumorization, and the incipient tumor cells that have higher autophagy capacity possibly are easier to survival in the hash microenvironment and accumulate more mutations to promote malignant progression.
doi:10.1186/2045-3701-3-35 pmid:24016776 pmcid:PMC3849558 fatcat:5wogjmzsz5avnooar7vl63wcyq

Weathering mechanism of red discolorations on Limestone object: a case study from Lingyan Temple, Jinan, Shandong Province, China

Jianrui Zha, Shuya Wei, Chuanchang Wang, Zhimin Li, Youzhen Cai, Qinglin Ma
2020 Heritage Science  
Red discolorations are an effloresce phenomenon detect on the surface of stone objects and considered as damage factor in both esthetic and conservation points of view. It is very difficult to remove and seldom report about their weathering mechanism. Recently, numerous of red discolorations have affected the limestone objects of Lingyan Temple in Shandong province, one of the most important building materials in China. In order to set up the appropriate conservation remedy, it is essential to
more » ... dentify the origin, characteristics, composition, and the formation process of red discolorations. Several analytical and investigation techniques, such as X-ray fluorescence spectroscopy (XRF), X-ray diffraction (XRD), Micro-Raman spectroscopy (Raman), Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM) equipped with energy dispersive spectrometer (EDS) were used to better understand the red discolorations. The results demonstrated that the discolorations on limestone surface have been caused by carbonic acid weathering process. The red discolorations consisted mainly of kutnohorite (Ca(Fe,Mg,Mn)CO 3 ) and iron oxides. They showed tabular, lamellar, and granular morphologies, which originated from the in situ carbonic acid weathering of kutnohorite. After rainfall, the Ca, Mg, Mn ions with relatively high solubility were primarily leached from carbonatite phases. It was resulting in the sedimentation of red iron oxides through a chemical reaction and physical adhesion. Based on those analyses, a chelating agent (ethylenediamine tetraacetic acid disodium salt) was chosen to remove theses red discolorations on the stone object. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
doi:10.1186/s40494-020-00394-z fatcat:u67tkropjna75ds2dgrek2kqfi

Massive molecular outflow and 100 kpc extended cold halo gas in the enormous Lyα nebula of QSO 1228+3128 [article]

Jianrui Li, Bjorn H. C. Emonts, Zheng Cai, J. Xavier Prochaska, Ilsang Yoon, Matthew D. Lehnert, Shiwu Zhang, Yunjing Wu, Jianan Li, Mingyu Li, Mark Lacy, Montserrat Villar-Martín
2021 arXiv   pre-print
., Reuland et al. 2003; Villar-Martín et al. 2003; Miley et al. 2006; Cai et al. , 2018)) .  ...  (e.g., Pâris et al. 2017) These QSOs contain Enormous Lyα Nebulae (ELANe) mapped by the Keck Cosmic Web Imager (KCWI) (Cai et al. 2019 ).  ... 
arXiv:2111.06409v1 fatcat:s7ksxycdovbz7oo4cc5a4uppki

Learning Image-adaptive 3D Lookup Tables for High Performance Photo Enhancement in Real-time

Hui Zeng, Jianrui Cai, Lida Li, Zisheng Cao, Lei Zhang
2020 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Cai L. Li and L. Zhang are with the Department of Computing, The Hong Kong Polytechnic University, Hong Kong. (email: {cshzeng, csjcai, cslli, cslzhang}@comp.polyu.edu.hk). • Z.  ... 
doi:10.1109/tpami.2020.3026740 pmid:32976094 fatcat:3eohl4sq5zhrlebvw2s5xe7ciq

Massive Molecular Outflow and 100 kpc Extended Cold Halo Gas in the Enormous Lyα Nebula of QSO 1228+3128

Jianrui Li, Bjorn H. C. Emonts, Zheng Cai, J. Xavier Prochaska, Ilsang Yoon, Matthew D. Lehnert, Shiwu Zhang, Yunjing Wu, Jianan Li, Mingyu Li, Mark Lacy, Montserrat Villar-Martín
2021 Astrophysical Journal Letters  
., Reuland et al. 2003; Villar-Martín et al. 2003; Miley et al. 2006; Cai et al. , 2018)) .  ...  These QSOs contain ELANe mapped by the Keck Cosmic Web Imager (KCWI; Cai et al. 2019 ).  ... 
doi:10.3847/2041-8213/ac390d fatcat:5ocbb5ht4vfd3moq44okwfh6za

GW26-e1582 The Relation between Glycosylated Hemoglobin and Severity of Coronary Artery Lesions in Patients with Coronary Heart Disease and Type 2 Diabetes Mellitus

Jianrui Zheng, Zhenda Zheng, Xing Shui, Dinghui Liu, Zheqi Wen, Chen Lin
2015 Journal of the American College of Cardiology  
GW26-e1582 The Relation between Glycosylated Hemoglobin and Severity of Coronary Artery Lesions in Patients with Coronary Heart Disease and Type 2 Diabetes Mellitus Jianrui Zheng, Zhenda Zheng, Xing Shui  ...  GW26-e3972 Non-alcoholic fatty liver disease affects the relationship between epicardium and coronary atherosclerotic heart disease He Cai, Shouyan Hao, Yang Zheng The first hospital of JILIN university  ... 
doi:10.1016/j.jacc.2015.06.524 fatcat:6tz6evlhhvbhppcn6mqn2zk7ue

GW26-e4557 Allocrytopine attenuates cardiac transmural dispersion of repolarization and protects ischemia reperfusion induced arrhythmias in rabbits

Yicheng Fu, Ying Zhao, Yu Zhang, Zhongqi Cai, Qing Dan, Xi Chen, Tong Yin, Yang Li
2015 Journal of the American College of Cardiology  
GW26-e4557 Allocrytopine attenuates cardiac transmural dispersion of repolarization and protects ischemia reperfusion induced arrhythmias in rabbits Yicheng Fu, Ying Zhao, Yu Zhang, Zhongqi Cai, Qing Dan  ...  e4600 Clinical effect of recombinant human brain natriuretic peptide combined with levosimendan on acute myocardial infarction complicated with heart failure Zhenda Zheng, Cailian Cheng, Dinghui Liu, Jianrui  ... 
doi:10.1016/j.jacc.2015.06.516 fatcat:b4xb3juzofebrdelhwmnkjpc24

GW26-e3913 Correlation between clopidogrel low response and CYP2C19 gene polymorphism in CAD patients after PCI

Shengli Yang, Yaqin Sun, Yong Yang
2015 Journal of the American College of Cardiology  
GW26-e1582 The Relation between Glycosylated Hemoglobin and Severity of Coronary Artery Lesions in Patients with Coronary Heart Disease and Type 2 Diabetes Mellitus Jianrui Zheng, Zhenda Zheng, Xing Shui  ...  GW26-e3972 Non-alcoholic fatty liver disease affects the relationship between epicardium and coronary atherosclerotic heart disease He Cai, Shouyan Hao, Yang Zheng The first hospital of JILIN university  ... 
doi:10.1016/j.jacc.2015.06.525 fatcat:ic7elyttyfc4xjl6yftq3robg4

GW26-e1478 Pridictive Value of Soluble Urokinase-type Plasminogen Activator Receptor in Patients with Acute Coronary Syndrome Undergoing Percutaneous Coronary Intervention

Zhenda Zheng, Cailian Cheng, Caihong Qu, Ruimin Dong, Min Wang, Lin Chen, Xiaoxian Qian
2015 Journal of the American College of Cardiology  
GW26-e1582 The Relation between Glycosylated Hemoglobin and Severity of Coronary Artery Lesions in Patients with Coronary Heart Disease and Type 2 Diabetes Mellitus Jianrui Zheng, Zhenda Zheng, Xing Shui  ...  GW26-e3972 Non-alcoholic fatty liver disease affects the relationship between epicardium and coronary atherosclerotic heart disease He Cai, Shouyan Hao, Yang Zheng The first hospital of JILIN university  ... 
doi:10.1016/j.jacc.2015.06.523 fatcat:mydrq3qkxrazvgfokdubhxbmru
« Previous Showing results 1 — 15 out of 36 results