检索结果-内蒙古大学图书馆

4th International conference on Machine Learning, image Processing, Network Security and data Sciences (MIND)

作者： Guo, Hanlin United Int Coll Div Culture & Creat Anshan 114000 Liaoning Peoples R China

ISBN: (纸本)9783031243660;9783031243677

The animation market structure continues to be optimized, and the animation economic cake continues to grow. Character image building is a very important work content, this part of the work is the most important part of the whole film and TVseries. Animation characters from the character, characteristics, character psychological activities, action and other aspects to show their own character image. All these features are innovative elements in film and television, which provide their own services for the development of film and television plot and make the development of film and television more visual. This paper studies the research on the image reconstruction of Internet of Things artificial intelligence and virtual reality technology in film and television character reconstruction, and points out the related content of the research on the image reconstruction of film and television character reconstruction. The test confirmed that artificial intelligence of the Internet of Things and virtual reality technology had excellent performance in the image reconstruction of film and television character role shaping.

关键词： Internet of things artificial intelligence Virtual reality technology Film and television character shaping image reconstruction research

来源：评论

学校读者我要写书评

暂无评论

A GENERAL FRAMEWORK FOR incomplete CROSS-MODAL RETRIEVAL WITH MISSING LABELS AND MISSING MODALITIES 47

A GENERAL FRAMEWORK FOR INCOMPLETE CROSS-MODAL RETRIEVAL WIT...

引用

47th IEEE International conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Li, Mingyang Huang, Shao-Lun Zhang, Lin Tsinghua Univ Tsinghua Berkeley Shenzhen Inst TBSI Shenzhen Peoples R China

ISBN: (纸本)9781665405409

Among various cross-modal retrieval methods, the supervised methods achieve the best performance by exploiting the semantic labels. However, in realistic applications, the data are not always complete with labels and full multi-modal data, which makes these methods hard to be used. In this paper, we propose a general framework for handling cross-modal retrieval tasks with both missing labels and missing modalities. To be more specific, in our framework we embed the data in each modality and labels all into a common feature space and maximize their correlation altogether. When labels or data in some modalities are missing, we can still maximize the correlation between the remaining data or labels. Combined with the label prediction and data reconstruction modules, our model can effectively extract useful information from the incomplete data for cross-modal retrieval tasks. In the extensive experiments, our model outperforms many other methods on different datasets, which proves the effectiveness and flexibility for handling incomplete data of our model.

关键词： Cross-modal Retrieval Multi-modal Learning Missing Modality

来源：评论

学校读者我要写书评

暂无评论

Accurate 3-D reconstruction Under IoT Environments and Its Applications to Augmented Reality

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2021年第3期17卷 2090-2100页

作者： Cao, Mingwei Zheng, Liping Jia, Wei Lu, Huimin Liu, Xiaoping Hefei Univ Technol Key Lab Knowledge Engn Big Data Minist Educ Hefei 230009 Anhui Peoples R China Hefei Univ Technol Anhui Prov Key Lab Ind Safety & Emergency Technol Hefei 230009 Peoples R China Hefei Univ Technol Sch Comp Sci & Informat Engn Hefei 230009 Peoples R China Kyushu Inst Technol Kitakyushu Fukuoka 8048550 Japan

With the remarkable development of sensor devices and the Internet of Things (IoT), today's researchers can easily know what changes have taken place in the real world by acquiring a 3-D model. Conversely, a large amount of image data promotes the development of perceptual computing technology. In this article, we focus on modeling 3-D scenes from the multisource image data obtained from the IoT with cameras. Although great progress has been made in 3-D reconstruction, it is still challenging to recover the 3-D model from IoT data because the captured images are usually noisy, incomplete, varying scale, and with repetitive structures or features. In this article, we propose an accurate 3-D reconstruction method under IoT environments for perceptual computing of the scene. This method consists of sparse, dense, and surface reconstruction processes, which can gradually recover high-quality geometric models from the image data and efficiently deal with various repetitive structures. By analyzing the reconstructed model, we can detect the changes of scenes. We evaluate the proposed method on the benchmark data sets (i.e., tanks and temples) and publicly available data sets(in which samples usually contain repeated structures, lighting change, and different scales). Experimental results show that the proposed method outperforms the state-of-the-art methods according to the standard evaluation metric. We also use our method to enhance the real scenes with virtual objects, thus producing promising results.

关键词： Three-dimensional displays image reconstruction Cameras Feature extraction Computational modeling Solid modeling Surface reconstruction Augmented reality Internet of Things (IoT) mixed Reality modeling 3-D reconstruction

来源：评论

学校读者我要写书评

暂无评论

The reconstruction, testing and improvement of a historical giant astronomical telescope 8

The reconstruction, testing and improvement of a historical ...

引用

2025 conference on Ultra-High-Definition Imaging Systems

作者： Osten, Wolfgang Pruss, Christof Herkommer, Alois Schober, Christian Treptow, Kevin Wischert, Michael Leue, Hans-Joachim Telescopium GmbH Koppel 25 D-28865 Lilienthal Germany Univ Stuttgart Inst Tech Opt Pfaffenwaldring 9 D-70569 Stuttgart Germany

ISBN: (纸本)9781510685260;9781510685277

Around 1800, the small German city Lilienthal housed the best telescopes of the time and became the founding place of the world's first astronomical society in which scientists from 24 European observatories together patternd the firmament according to the lack of a planet between Mars and Jupiter. Lilienthal and his observatories were already in the late 18th and early 19th century sites of the encounter and the scientific exchange of numerous high-ranking scholars such as Bessel, Gauss, Olbers, and Harding. The central instrument was the 27 foot Giant Telescope where important discoveries were made. However, this first scientific institution of European format was to fall victim to a fire in 1813, as a result of which almost the entire scientific heritage of the founder, Johann Hieronymus Schroeter, was destroyed. The detailed reconstruction of the Schroeter Giant Telescope was only carried out in 2015 through exclusively private commitment from Lilienthal citizens and donors from Hamburg and Bremen. One of the most important contributions of the group around Schroeter was the development and implementation of a new technology for the fabrication of telescope mirrors. In contrast to those made by Herschel, these mirrors are characterized by a high surface quality, the reflective shine of which did not fade even after decades of use. In this paper we show a comparison between the image quality of the original spherical mirror made from a copper-tin-arsenic alloy with an aperture of 50 cm and a focal length of 27 foot that was fabricated in 1793 and a new mirror made from Schott borosilicate glass Supremax 33 with the same design data. Furthermore we show some optical concepts how to mimic and compare the historical image quality to the state of the art and to improve the imaging properties of the modern telescope.

关键词： astronomical telescopes mirror fabrication image quality interferometry Ronchi test aberrations aberration correction

来源：评论

学校读者我要写书评

暂无评论

Med-DANet V2: A Flexible Dynamic Architecture for Efficient Medical Volumetric Segmentation

Med-DANet V2: A Flexible Dynamic Architecture for Efficient ...

引用

IEEE/CVF Winter conference on Applications of Computer Vision (WACV)

作者： Shen, Haoran Zhang, Yifu Wang, Wenxuan Chen, Chen Liu, Jing Song, Shanshan Li, Jiangyun Univ Sci & Technol Beijing Sch Automat & Elect Engn Beijing Peoples R China Univ Cent Florida Ctr Res Comp Vis Orlando FL 32816 USA Chinese Acad Sci Inst Automat Natl Lab Pattern Recognit Beijing Peoples R China

ISBN: (纸本)9798350318920;9798350318937

Recent works have shown that the computational efficiency of 3D medical image (e.g. CT and MRI) segmentation can be impressively improved by dynamic inference based on slice-wise complexity. As a pioneering work, a dynamic architecture network for medical volumetric segmentation (i.e. Med-DANet [44]) has achieved a favorable accuracy and efficiency trade-off by dynamically selecting a suitable 2D candidate model from the pre-defined model bank for different slices. However, the issues of incomplete data analysis, high training costs, and the two-stage pipeline in Med-DANet require further improvement. To this end, this paper further explores a unified formulation of the dynamic inference framework from the perspective of both the data itself and the model structure. For each slice of the input volume, our proposed method dynamically selects an important foreground region for segmentation based on the policy generated by our Decision Network and Crop Position Network. Besides, we propose to insert a stage-wise quantization selector to the employed segmentation model (e.g. U-Net) for dynamic architecture adapting. Extensive experiments on BraTS 2019 and 2020 show that our method achieves comparable or better performance than previous state-of-the-art methods with much less model complexity. Compared with previous methods Med-DANet and TransBTS with dynamic and static architecture respectively, our framework improves the model efficiency by up to nearly 4.1 and 17.3 times with comparable segmentation results on BraTS 2019. Code will be available at https://***/ RubicsXuan/Med- DANet.

关键词： Algorithms Applications Biomedical / healthcare / medicine image recognition and understanding

来源：评论

学校读者我要写书评

暂无评论

reconstruction of Medical images from Sparse data: A Deep Learning Approach 5

Reconstruction of Medical Images from Sparse Data: A Deep Le...

引用

5th International conference on Mobile Computing and Sustainable Informatics, ICMCSI 2024

作者： Saravanan, S. Bruntha, P.Malin Joseph S, Iwin Thanakumar Subramanian, Suresh Sundar, G. Naveen Narmadha, D. VIT School of Design Vellore Institute of Technology Department of Multimedia Vellore India Karunya Institute of Technology and Sciences Department of Electronics and Communication Engineering Coimbatore India Koneru Lakshmaiah Education Foundation Department of CSE Andhrapradesh Vaddeswaram India Saveetha School of Engineering Saveetha Institute of Medical and Technical Science Department of Electronics and Communication Engineering Chennai India Karunya Institute of Technology and Sciences Department of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798350395235

For diagnosis, medical imaging technology plays a significant role in today's scenario. Accordingly, enormous amounts of medical data are generated daily on research centers and hospitals. Storing the medical images needs a lot of space where in 3D (Three dimensional) medical data acquires a huge space over it. All the medical image details need to be stored digitally for efficient diagnosis. The Digital Imaging and Communications in Medicine (DICOM) standard was established as the global norm for digitally storing patient images from diverse medical imaging modalities. Efficient compression over medical images is required to maintain efficacy in storage space. Compression over images like JPEG, PNG, etc. may not be efficient in bringing an effective lossless image compression in all the types of imaging modalities. Hence, an efficient methodology of image compression over medical images are required in today's scenario. This research study compares the performance of different methods of compression standard for medical images with the proposed method called DL Based Compressive sensing on Medical image compression. The primary aim of Compressive Sensing (CS) is to achieve compression of a sparse signal at a sampling rate below the Nyquist rate. Results are analyzed using performance metrics like PSNR, MSE, BPP and CR. And it proves that the proposed algorithm achieves a higher PSNR value and states that it is efficient in bringing a lossless medical image compression. © 2024 IEEE.

关键词： image compression

来源：评论

学校读者我要写书评

暂无评论

Features Based reconstruction Technique for Torn Documents

Features Based Reconstruction Technique for Torn Documents

引用

2023 International conference on Green Energy, Computing and Intelligent Technology, GEn-CITy 2023

作者： Saleh, Marwan D. Salih, N.D. Isa, Wan Noorshahida Mohd Electronics and Computer Science University of Southampton Malaysia Johor Malaysia Faculty of Computing and Informatics Multimedia University Selangor Cyberjaya Malaysia

ISBN: (纸本)9781839539596

The reconstruction of torn documents that hold sensitive information or crucial historical data is of great significance in many domains. An efficient technique for hand-torn document restoration is presented in this paper. Pre-processing, feature extraction, and matching three main phases of this technique. The reconstruction is carried out by matching pairwise the feature vectors of the fragments. Using a database consists of 120 torn pages taken from books, newspapers and magazines, the performance of this technique has been measured using simple but effective criteria called error rate. The algorithm was able to accurately reconstruct 91% of the documents, with the remaining being partially reconstructed. © The Institution of Engineering & Technology 2023.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Flemme: A Flexible and Modular Learning Platform for Medical images

Flemme: A Flexible and Modular Learning Platform for Medical...

引用

2024 IEEE International conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Zhang, Guoqing Yang, Jingyun Li, Yang Tsinghua University Tsinghua Shenzhen International Graduate School Shenzhen China Pengcheng Laboratory Shenzhen China The Shenzhen Key Laboratory of Ubiquitous Data Enabling SIGS Tsinghua China

ISBN: (纸本)9798350386226

With the rapid development of computer vision and the emergence of powerful network backbones and architectures, the application of deep learning in medical imaging has become increasingly significant. Unlike natural images, medical images lack huge volumes of data but feature more modalities, making it difficult to train a general model that has satisfactory performance across various datasets. In practice, practitioners often suffer from manually creating and testing models combining independent backbones and architectures, which is a laborious and time-consuming process. We propose Flemme, a FLExible and Modular learning platform for MEdical images. Our platform separates encoders from the model architectures so that different models can be constructed via various combinations of supported encoders and architectures. We construct encoders using building blocks based on convolution, transformer, and state-space model (SSM) to process both 2D and 3D image patches. A base architecture is implemented following an encoder-decoder style, with several derived architectures for image segmentation, reconstruction, and generation tasks. In addition, we propose a general hierarchical architecture incorporating a pyramid loss to optimize and fuse vertical features. Experiments demonstrate that this simple design leads to an average improvement of 5.60% in Dice score and 7.81% in mean interaction of units (mIoU) for segmentation models, as well as an enhancement of 5.57% in peak signal-to-noise ratio (PSNR) and 8.22% in structural similarity (SSIM) for reconstruction models. We further utilize Flemme as an analytical tool to assess the effectiveness and efficiency of various encoders across different tasks. Code is available at https://***/wlsdzyzl/flemme. © 2024 IEEE.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

High Dynamic Range image Recovery by Use of Lens Flare Events Detection Algorithm

High Dynamic Range Image Recovery by Use of Lens Flare Event...

引用

2023 International conference on Electronics, Information, and Communication, ICEIC 2023

作者： Chang, Bobaro Ryu, Hyunsurk Lee, Hyuk-Jae Seoul National University Department of Electrical and Computer Engineering Seoul Korea Republic of

ISBN: (纸本)9798350320213

Event cameras are novel vision sensors that asynchronously output per-pixel events by measuring brightness changes. Event cameras have advantages such as high speed and high dynamic range compared to conventional cameras. To leverage the advantages and apply prior algorithms, event-based image reconstruction has been developed. With the development of neural networks, state-of-the-art reconstruction methods are introduced. However, high dynamic range reconstructions still suffer from lens flare-based artefacts, which makes the intensity estimation incorrect. To address this problem, this work presents a computational approach to detect ill-posed events caused by lens flare. Under the examination that lens flare elements of event cameras mainly consist of glow and starburst, we derive two bivariate Gaussian distributions from targets in the compressed stream of events. By operating convolution, the detector reduces the sparsity of events, which makes the surface fitting more precise. We show that the proposed method effectively eliminates dark stain in high dynamic range reconstruction, while preserving detail on the other region at the same time. © 2023 IEEE.

关键词： Event camera Gaussian distribution high dynamic range image reconstruction lens flare

来源：评论

学校读者我要写书评

暂无评论

Unstructured Road image Enhancement Method in Amblyopia Environment Based on Improved Dark Channel Prior 8

Unstructured Road Image Enhancement Method in Amblyopia Envi...

引用

8th CAA International conference on Vehicular Control and Intelligence, CVCI 2024

作者： Xue, Pengyu Yang, Chen Cheng, Yuejun Feng, Ping Wang, Hongliang Pi, Dawei School of Mechanical Engineering Nanjing University of Science and Technology Jiangsu Nanjing China

ISBN: (纸本)9798331504892

Perception and recognition of roadable areas is an important part of intelligent driving. Most of the existing researches on road perception methods are limited to structured roads. However, in the complex field environment, the difficulty of identifying roadable area surges, so it is necessary to develop a road perception method that adapts to the field environment. Based on existing sensor equipment, through the information fusion of multi-sensor data, driving area in free space can be detected and identified, and the driving efficiency and safety of vehicles in amblyopia environment can be improved. This paper takes intelligent vehicle as the object to carry out the research of environmental perception enhancement. An image enhancement algorithm for amblyopia environment was designed, and the transmission accuracy fitting experiment was carried out by improving the dark channel prior theory. The fitting results were applied to the smoke image to be processed by combining haze image degradation model, and the smoke image reconstruction was carried out to ensure the high definition and fidelity of the image in the amblyopia environment. © 2024 IEEE.

关键词： Information fusion

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：