Xuanyu Zhang

Dr. Xuanyu Zhang

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 April 2024 in CAAI Transactions on Intelligence Technology
Verified
Journal Article
Open access
PDF file
Chunwei TianChunwei Tian
Chunwei Tian
XZXuanyu Zhang
Xuanyu Zhang
QZQi Zhang
Qi Zhang
MYMingming Yang
Mingming Yang
ZJZhaojie Ju
Zhaojie Ju
Convolutional neural networks depend on deep network architectures to extract accurate information for image super‐resolution. However, obtained information of these convolutional neural networks cannot completely express predicted high‐quality images for complex scenes. A dynamic network for image super‐resolution (DSRNet) is presented, which contains a residual enhancement block, wide enhancement block, feature refinement block and construction block. The residual enhancement block is composed of a residual enhanced architecture to facilitate hierarchical features for image super‐resolution. To enhance robustness of obtained super‐resolution model for complex scenes, a wide enhancement block achieves a dynamic architecture to learn more robust information to enhance applicability of an obtained super‐resolution model for varying scenes. To prevent interference of components in a wide enhancement block, a refinement block utilises a stacked architecture to accurately learn obtained features. Also, a residual learning operation is embedded in the refinement block to prevent long‐term dependency problem. Finally, a construction block is responsible for reconstructing high‐quality images. Designed heterogeneous architecture can not only facilitate richer structural information, but also be lightweight, which is suitable for mobile digital devices. Experimental results show that our method is more competitive in terms of performance, recovering time of image super‐resolution and complexity. The code of DSRNet can be obtained at https://github.com/hellloxiaotian/DSRNet .

6 November 2023 in IEEE Transactions on Intelligent Vehicles
Verified
Journal Article
Closed access
Chunwei TianChunwei Tian
Chunwei Tian
XZXuanyu Zhang
Xuanyu Zhang
XLXu Liang
Xu Liang
Bo LiBo Li
Bo Li
Yougang ShanghaiYougang Shanghai
Yougang Shanghai
SZShichao Zhang
Shichao Zhang
Deep convolutional neural networks (CNNs) can improve recognition rate in license plate to improve traffic. However, these methods may refer to big computational costs and a lot of parameters. In this paper, we propose a knowledge distillation with a fast CNN for license plate detection (KDNet). KDNet uses knowledge distillation to guide a CNN to optimize parameters and quickly obtain a detector for license plate. To overcome naive effect of local information, a non-local similarity mechanism is used into a CNN to enhance effect of global information for extracting salient information in license plate detection. Experimental results that this proposed KDNet is superior to detection speed for license plate. The code of KDNet can be obtained at https://github.com/hellloxiaotian/KDNet .

3 October 2022 in arXiv
Verified
Preprint
Open access
PDF file
Chunwei TianChunwei Tian
Chunwei Tian
XZXuanyu Zhang
Xuanyu Zhang
Jerry Chun-Wei LinJerry Chun-Wei Lin
Jerry Chun-Wei Lin
WZWangmeng Zuo
Wangmeng Zuo
YZYanning Zhang
Yanning Zhang
CLChia-Wen Lin
Chia-Wen Lin
Single image super-resolution (SISR) has played an important role in the field of image processing. Recent generative adversarial networks (GANs) can achieve excellent results on low-resolution images with small samples. However, there are little literatures summarizing different GANs in SISR. In this paper, we conduct a comparative study of GANs from different perspectives. We first take a look at developments of GANs. Second, we present popular architectures for GANs in big and small samples for image applications. Then, we analyze motivations, implementations and differences of GANs based optimization methods and discriminative learning for image super-resolution in terms of supervised, semi-supervised and unsupervised manners. Next, we compare performance of these popular GANs on public datasets via quantitative and qualitative analysis in SISR. Finally, we highlight challenges of GANs and potential research points for SISR.

8 November 2025 in Remote Sensing
Verified
Journal Article
Open access
PDF file
XDXiaokun Ding
Xiaokun Ding
XZXuanyu Zhang
Xuanyu Zhang
SSShangzhen Song
Shangzhen Song
BLBo Li
Bo Li
LHLe Hui
Le Hui
YDYuchao Dai
Yuchao Dai
Cross-view geo-localization allows an agent to determine its own position by retrieving the same scene from images taken from dramatically different perspectives. However, image matching and retrieval face significant challenges due to substantial viewpoint differences, unknown orientations, and considerable geometric distribution disparities between cross-view images. To this end, we propose a cross-view geo-localization framework based on novel view synthesis that generates pseudo aerial-view images from given street-view scenes to reduce the view discrepancies, thereby improving the performance of cross-view geo-localization. Specifically, we first employ 3D Gaussian splatting to generate new aerial images from the street-view image sequence, where COLMAP is used to obtain initial camera poses and sparse point clouds. To identify optimal matching viewpoints from reconstructed 3D scenes, we design an effective camera pose estimation strategy. By increasing the tilt angle between the photographic axis and the horizontal plane, the geometric consistency between the newly generated aerial images and the real ones can be improved. After that, the DINOv2 is employed to design a simple yet efficient mixed feature enhancement module, followed by the InfoNCE loss for cross-view geo-localization. Experimental results on the KITTI dataset demonstrate that our approach can significantly improve cross-view matching accuracy under large viewpoint disparities and achieve state-of-the-art localization performance.

25 August 2023 in Mathematics
Verified
Journal Article
Open access
PDF file
XZXuanyu Zhang
Xuanyu Zhang
Chunwei TianChunwei Tian
Chunwei Tian
QZQi Zhang
Qi Zhang
HGHong Seng Gan
Hong Seng Gan
TCTongtong Cheng
Tongtong Cheng
MIMohd Asrul Hery Ibrahim
Mohd Asrul Hery Ibrahim
Due to COVID-19, online education has become an important tool for teachers to teach students. Also, teachers depend on a multimedia teaching system (platform) to finish online education. However, interacted images from a multimedia teaching system may suffer from noise. To address this issue, we propose a lightweight image denoising network (LIDNet) for multimedia teaching systems. A parallel network can be used to mine complementary information. To achieve an adaptive CNN, an omni-dimensional dynamic convolution fused into an upper network can automatically adjust parameters to achieve a robust CNN, according to different input noisy images. That also enlarges the difference in network architecture, which can improve the denoising effect. To refine obtained structural information, a serial network is set behind a parallel network. To extract more salient information, an adaptively parametric rectifier linear unit composed of an attention mechanism and a ReLU is used into LIDNet. Experiments show that our proposed method is effective in image denoising, which can also provide assistance for multimedia teaching systems.

1 May 2021 in IOP Conference Series: Earth and Environmental Science
Verified
Conference Paper
Open access
PDF file
DPDong Peng
Dong Peng
ZSZhu Shichao
Zhu Shichao
YXYan Xin
Yan Xin
XZXuanyu Zhang
Xuanyu Zhang
In view of the management characteristics of highway engineering, it integrates BIM, mobile Internet, cloud computing, Internet of things and other technologies to study and establish a fine and collaborative management platform for project progress, quality and safety on the basis of mobile terminal application. This paper realize the dynamic control of project construction resources with progress as the main line and the collaborative management of projects with safety and quality as the core. It can meet the application analysis demands of all parties involved in the construction of highway for the whole service data, and give full play to the convenience of mobile terminals, and carry out task coordination and mobile operation of various management work in time through the mobile terminal anytime and anywhere. At the same time, the data of the mobile terminal is associated with the structure through BIM model, which can be statistically analyzed and visually displayed, and realize the collaborative association of various business data, thus greatly improving the refined management level of highway engineering construction.