检索结果-内蒙古大学图书馆

A survey on image and video stitching

virtual Reality & Intelligent Hardware 2019年第1期1卷 55-83页

作者： Wei LYU Zhong ZHOU Lang CHEN Yi ZHOU State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and EngineeringBeihang UniversityBeijing 100191China

Image/video stitching is a technology for solving the field of view(FOV)limitation of images/*** stitches multiple overlapping images/videos to generate a wide-FOV image/video,and has been used in various fields such as sports broadcasting,video surveillance,street view,and *** survey reviews image/video stitching algorithms,with a particular focus on those developed in recent *** stitching first calculates the corresponding relationships between multiple overlapping images,deforms and aligns the matched images,and then blends the aligned images to generate a wide-FOV image.A seamless method is always adopted to eliminate such potential flaws as ghosting and blurring caused by parallax or objects moving across the overlapping *** stitching is the further extension of image *** usually stitches selected frames of original videos to generate a stitching template by performing image stitching algorithms,and the subsequent frames can then be stitched according to the *** stitching is more complicated with moving objects or violent camera movement,because these factors introduce jitter,shakiness,ghosting,and *** detection technique is usually combined into stitching to eliminate ghosting and blurring,while video stabilization algorithms are adopted to solve the jitter and *** paper further discusses panoramic stitching as a special-extension of image/video *** stitching is currently the most widely used application in *** survey reviews the latest image/video stitching methods,and introduces the fundamental principles/advantages/weaknesses of image/video stitching ***/video stitching faces long-term challenges such as wide baseline,large parallax,and low-texture problem in the overlapping *** technologies may present new opportunities to address these issues,such as deep learning-based semantic correspondence,and 3D image ***,

关键词： Image stitching Video stitching Panoramic stitching Registration Alignment Mesh optimization Deep learning 3D stitching

来源：评论

学校读者我要写书评

暂无评论

Efficient Low-Resolution Face Recognition via Bridge Distillation

arXiv

引用

arXiv 2024年

作者： Ge, Shiming Zhao, Shengwei Li, Chenyu Zhang, Yu Li, Jia The Institute of Information Engineering Chinese Academy of Sciences Beijing100095 China School of Cyber Security University of Chinese Academy of Sciences Beijing100049 China SenseTime Group Limited 100084 China The State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University China The Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing100191 China The Peng Cheng Laboratory Shenzhen518055 China

Face recognition in the wild is now advancing towards light-weight models, fast inference speed and resolution-adapted capability. In this paper, we propose a bridge distillation approach to turn a complex face model pretrained on private high-resolution faces into a light-weight one for low-resolution face recognition. In our approach, such a cross-dataset resolution-adapted knowledge transfer problem is solved via two-step distillation. In the first step, we conduct cross-dataset distillation to transfer the prior knowledge from private high-resolution faces to public high-resolution faces and generate compact and discriminative features. In the second step, the resolution-adapted distillation is conducted to further transfer the prior knowledge to synthetic low-resolution faces via multi-task learning. By learning low-resolution face representations and mimicking the adapted high-resolution knowledge, a light-weight student model can be constructed with high efficiency and promising accuracy in recognizing low-resolution faces. Experimental results show that the student model performs impressively in recognizing low-resolution faces with only 0.21M parameters and 0.057MB memory. Meanwhile, its speed reaches up to 14,705, 934 and 763 faces per second on GPU, CPU and mobile phone, respectively. Copyright © 2024, The Authors. All rights reserved.

关键词： Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Exploiting aliasing for manga restoration

arXiv

引用

arXiv 2021年

作者： Xie, Minshan Xia, Menghan Wong, Tien-Tsin Department of Computer Science and Engineering Chinese University of Hong Kong Hong Kong Shenzhen Key Laboratory of Virtual Reality and Human Interaction Technology SIAT CAS

As a popular entertainment art form, manga enriches the line drawings details with bitonal screentones. However, manga resources over the Internet usually show screentone artifacts because of inappropriate scanning/rescaling resolution. In this paper, we propose an innovative twostage method to restore quality bitonal manga from degraded ones. Our key observation is that the aliasing induced by downsampling bitonal screentones can be utilized as informative clues to infer the original resolution and screentones. First, we predict the target resolution from the degraded manga via the Scale Estimation Network (SENet) with spatial voting scheme. Then, at the target resolution, we restore the region-wise bitonal screentones via the Manga Restoration Network (MR-Net) discriminatively, depending on the degradation degree. Specifically, the original screentones are directly restored in pattern-identifiable regions, and visually plausible screentones are synthesized in pattern-agnostic regions. Quantitative evaluation on synthetic data and visual assessment on real-world cases illustrate the effectiveness of our method. Copyright © 2021, The Authors. All rights reserved.

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

Toward Lung Ultrasound Automation: Fully Autonomous Robotic Longitudinal and Transverse Scans Along Intercostal Spaces

IEEE Transactions on Medical Robotics and Bionics

引用

IEEE Transactions on Medical Robotics and Bionics 2025年第2期7卷 768-781页

作者： Long Lei Yingbai Hu Zixing Jiang Juzheng Miao Xiao Luo Yu Zhang Qiong Wang Shujun Wang Zheng Li Pheng-Ann Heng Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China Multi-Scale Medical Robotics Centre Ltd. The Chinese University of Hong Kong Hong Kong China Department of Surgery The Chinese University of Hong Kong Hong Kong China Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China Department of Biomedical Engineering The Hong Kong Polytechnic University Hong Kong China Department of Surgery Chow Yuk Ho Technology Centre for Innovative Medicine Li Ka Shing Institute of Health Sciences and Multiscale Medical Robotics Centre Ltd. The Chinese University of Hong Kong Hong Kong China

Lung ultrasound scanning is essential for diagnosing lung diseases. The scan effectiveness critically depends on both longitudinal and transverse scans through intercostal spaces to reduce rib shadowing interference, as well as maintaining the probe perpendicular to pleura for pathological artifact generation. Achieving this level of scan quality often depends heavily on the experience of doctors. Robotic ultrasound scanning shows promise, but currently lacks a direct path planning method for intercostal scanning, and probe orientation does not consider imaging differences between lungs and solid organs. In this paper, we aim to fully automate two fundamental operations in lung ultrasound scanning: longitudinal and transverse scans. We propose pioneering path planning methods along intercostal spaces and innovative solutions for adaptive probe posture adjustment using real-time pleural line feedback, specifically addressing the unique characteristics of lung ultrasound scanning. This ensures the acquisition of high-quality, diagnostically meaningful ultrasound images. In addition, we develop a robotic lung ultrasound system to validate the proposed methods. Extensive experimental results on two volunteers and a chest phantom confirm the efficacy of our methods, and demonstrate the system’s feasibility in automated lung ultrasound examinations. Our work lays a solid foundation for automated robotic complete lung scanning.

关键词： Ultrasonic imaging Lungs Probes Robots Robot kinematics Medical robotics Ribs Medical services Surface morphology Computed tomography

来源：评论

学校读者我要写书评

暂无评论

Optimization of Load Balancing Algorithm for virtual Machine Dynamic Migration under Mobile Cloud Computing

引用

China Communications 2020年第6期17卷 237-245页

作者： Weijin Jiang Fang Ye Shengjie Yang Wei Liu Xiaoliang Liu Sijian Lv School of Computer Science and Technology Hunan University of Technology and BusinessChangsha 410205China Key Laboratory of Hunan Province for New Retail Virtual Reality Technology Hunan University of Technology and BusinessChangsha 410205China School of Computer Science and Technology Wuhan University of TechnologyWuhan 430073China

As one of the key technologies of cloud computing,the virtualization technology can virtualize all kinds of resources and integrate them into the unified planning of the cloud computing management *** migration of virtual machines is one of the important technologies of virtual machine ***,there are still many deficiencies in the implementation of load balancing by virtual machine dynamic migration in cloud *** triggering strategy thresholds are mostly *** there is an instantaneous peak,it will cause migration,which will cause a waste of *** order to solve this problem,based on improving the dynamic migration framework,this paper proposes node selection optimization algorithm and node load balancing strategy and designs a prediction module,which uses a one-time smooth prediction to avoid the shortcoming of peak load *** simulation experiments and conclusions analysis results show that the fusion algorithm has performance advantages obvious.

关键词： cloud computing dynamic migration load balance node selection load value prediction

来源：评论

学校读者我要写书评

暂无评论

Learning Open Set Network with Discriminative Reciprocal Points 16th

Learning Open Set Network with Discriminative Reciprocal Poi...

引用

16th European Conference on computer Vision, ECCV 2020

作者： Chen, Guangyao Qiao, Limeng Shi, Yemin Peng, Peixi Li, Jia Huang, Tiejun Pu, Shiliang Tian, Yonghong Department of Computer Science and Technology Peking University Beijing China State Key Laboratory of Virtual Reality Technology and Systems SCSE Beihang University Beijing China Peng Cheng Laboratory Shenzhen China Hikvision Research Institute Hangzhou China

ISBN: (纸本)9783030585792

Open set recognition is an emerging research area that aims to simultaneously classify samples from predefined classes and identify the rest as ‘unknown’. In this process, one of the key challenges is to reduce the risk of generalizing the inherent characteristics of numerous unknown samples learned from a small amount of known data. In this paper, we propose a new concept, Reciprocal Point, which is the potential representation of the extra-class space corresponding to each known category. The sample can be classified to known or unknown by the otherness with reciprocal points. To tackle the open set problem, we offer a novel open space risk regularization term. Based on the bounded space constructed by reciprocal points, the risk of unknown is reduced through multi-category interaction. The novel learning framework called Reciprocal Point Learning (RPL), which can indirectly introduce the unknown information into the learner with only known classes, so as to learn more compact and discriminative representations. Moreover, we further construct a new large-scale challenging aircraft dataset for open set recognition: Aircraft 300 (Air-300). Extensive experiments on multiple benchmark datasets indicate that our framework is significantly superior to other existing approaches and achieves state-of-the-art performance on standard open set benchmarks. © 2020, Springer Nature Switzerland AG.

关键词： Aircraft

来源：评论

学校读者我要写书评

暂无评论

EPI-Guided Cost Construction Network for Light Field Disparity Estimation

EPI-Guided Cost Construction Network for Light Field Dispari...

引用

IEEE computer Society Conference on computer Vision and Pattern Recognition Workshops (CVPRW)

作者： Tun Wang Rongshan Chen Ruixuan Cong Da Yang Zhenglong Cui Fangping Li Hao Sheng State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing P.R. China Beihang Hangzhou Innovation Institute Yuhang Xixi Octagon City Hangzhou P.R. China Faculty of Applied Sciences Macao Polytechnic University Macao SAR P.R. China

Recent learning-based light field (LF) disparity estimation methods construct cost volume by sequentially shifting each sub-aperture image (SAI) with a series of predefined offsets. They only use the visual information of SAIs and lose the geometry of LF. In this paper, we design a simple network that can cleverly integrate EPI features with cost volume to estimate the disparity. Firstly, we propose an efficient EPI extraction module to use abundant line characteristics. Secondly, we offer an EPI-Cost volume construction module that can create volume guided by the EPI line and the color consistency of images. Finally, after completing it, we adopt an inter-volume fusion module to considerably correlate the validity of EPI lines in both directions. Experimental results show the proposed method achieves state-of-the-art performance in the quantitative and qualitative evaluation of the UrbanLF-Syn dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PcdGS: A point cloud densification method for Gaussian Splatting

引用

computers & Graphics 2025年 130卷

作者： Jing Zhao Chenxi Li Shuhui Zhang Feng Wang Ying Yin Yuhan Liu School of Information Science and Engineering Yanshan University Qinhuangdao 066000 Hebei China Hebei Key Laboratory of Computer Virtual Technology and System Integration Qinhuangdao Hebei China

Novel view synthesis (NVS) from limited observations continues to be an important and persistent challenge. Currently, methods based on Neural Radiance Fields (NeRF) are not very efficient, and although methods based on 3D Gaussian Splatting (3DGS) outperform NeRF in terms of rendering quality and speed, they still lack sufficient details. To tackle these issues, we introduce a new method for densifying the point cloud, which enhances the rendering effect of 3DGS-based techniques under sparse view conditions. First, we introduce a mask-based densification technique to improve rendering details under limited input views. Second, We propose a monocular pixel depth-based mapping method that leverages a pre-trained model to predict depth, effectively normalizing point locations within the resulting point cloud. Lastly, we implement a filtering method based on depth and RGB color to minimize noise introduced by the additional data. We conduct comparative experiments on the LLFF, Mip-NeRF 360, and Blender datasets, demonstrating that our method outperforms existing approaches in both evaluation metrics and visual quality, thereby validating its effectiveness.

关键词： 3D reconstruction

来源：评论

学校读者我要写书评

暂无评论

Separating Content and Style for Unsupervised Image-to-Image Translation

arXiv

引用

arXiv 2021年

作者： Liu, Yunfei Wang, Haofei Yue, Yang Lu, Feng State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Peng Cheng Laboratory Shenzhen China Beihang University Beijing China

Unsupervised image-to-image translation aims to learn the mapping between two visual domains with unpaired samples. Existing works focus on disentangling domain-invariant content code and domain-specific style code individually for multimodal purposes. However, less attention has been paid to interpreting and manipulating the translated image. In this paper, we propose to separate the content code and style code simultaneously in a unified framework. Based on the correlation between the latent features and the high-level domain-invariant tasks, the proposed framework demonstrates superior performance in multimodal translation, interpretability and manipulation of the translated image. Experimental results show that the proposed approach outperforms the existing unsupervised image translation methods in terms of visual quality and diversity. Code and data have been released at https://***/DreamtaleCore/SCS-UIT. Copyright © 2021, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Meta-RetinaNet for Few-shot Object Detection 31

Meta-RetinaNet for Few-shot Object Detection

引用

31st British Machine Vision Conference, BMVC 2020

作者： Li, Shaoqi Song, Wenfeng Li, Shuai Hao, Aimin Qin, Hong State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China Peng Cheng Laboratory Shenzhen518055 China Department of Computer Science Stony Brook University Stony Brook United States

Few shot object detection (FSD) is gaining popularity, enhanced by the deep learning methods in recent years. Meanwhile, meta-learning has achieved great success in few-shot image classification benefitting from its adaptive capability corresponding to a suite of tasks. Yet, most object detection models are based on deep neural networks (DNNs), and they are prone to the overfitting problem due to limited samples available during training. To adapt the learned prior knowledge more effectively to new tasks, this paper proposes a novel Meta-RetinaNet for FSD, which avoids a biased meta-learner and improves its generalization ability. It employs a Meta Coefficient Learner (MCL) trained by the Balanced Loss (BL) to augment the DNNs. Specifically, the MCL adapts to tasks by the product of pre-trained convolution weights and coefficient vectors densely for all the convolutional layers, such that it could adequately transfer the learned knowledge to new tasks (while overcoming the overfitting problem) by training fewer parameters. In addition, the BL expedites the training of a Meta-RetinaNet by balancing the performance of a host of tasks, and it also retains stable performance for new tasks. Our experiments showcase the effectiveness of our method, which achieves the state-of-the-art performance on the multiple settings of Pascal VOC and COCO datasets. © 2020. The copyright of this document resides with its authors. It may be distributed unchanged freely in print or electronic forms.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：