检索结果-内蒙古大学图书馆

16th IEEE/ACS international conference on computer Systems and Applications (AICCSA)

作者： Richter, Edward Raettig, Ryan Mack, Joshua Valancius, Spencer Unal, Burak Akoglu, Ali Univ Arizona Elect & Comp Engn Tucson AZ 85721 USA

ISBN: (数字)9781728150529

ISBN: (纸本)9781728150529

Shadows can have a negative effect on the ability of computer vision techniques for object detection, tracking, and recognition. therefore, ability to remove shadows and byproducts of illumination is an important problem to enable effective object recognition actions. As applications move into levels of higher information extraction and higher required processing speeds, efficient and sophisticated shadow detection and removal becomes even more necessary. In this study we propose a shadow removal method, parallelize using a Tesla P100 GPU, and achieve a speedup of 21.67x on an 18 megapixel (MP) resolution image compared to the same method implemented in Matlab.

关键词： Convolution Kernel graphics processing units Gray-scale Mathematical model computer vision Histograms

来源：评论

学校读者我要写书评

暂无评论

Online Panorama Image Generation for a Disaster Rescue Vehicle 16

Online Panorama Image Generation for a Disaster Rescue Vehic...

引用

16th international conference on Ubiquitous Robots (UR)

作者： Sovetov, Kanat Kim, Jun-Sik Kim, Doik UST KIST Sch Div Nano & Informat Technol Daejeon South Korea Korea Inst Sci & Technol Robot & Media Inst Ctr Intelligent & Interact Robot Seoul South Korea

ISBN: (纸本)9781728132327

We introduce our online panorama generation system for armored rescue vehicles in an extreme disaster. To make the whole system fireproof and heatproof, all the cameras are embedded around the vehicle body, and this makes the panorama stitching suffer from binocular parallax. Because vehicle and remote operators should observe its current surroundings, the panorama image should be generated online in an acceptable time. We explain how to calculate the initial alignment of input camera image on the cylindrical canvas in advance, and how to merge the overlapped image by removing the binocular parallax. Structures for the data transfer is also discussed. Experiments show that the current implementation can generate and share the panoramic image with remote operators successfully.

关键词： Cameras Rendering (computer graphics) Robot vision systems Calibration Feature extraction"

来源：评论

学校读者我要写书评

暂无评论

Deep coupling neural network for robust facial landmark detection

引用

computerS & graphics-UK 2019年 82卷 286-294页

作者： Wu, Wenyan Wu, Xingzhe Cai, Yici Zhou, Qiang Tsinghua Univ Beijing Natl Res Ctr Informat Sci & Technol BNRis Dept Comp Sci & Technol Beijing Peoples R China Beihang Univ Coll Software Beijing Peoples R China

Facial landmark detection aims at locating a sparse set of fiducial facial key-points. Two significant issues (i.e., Intra-Dataset Variation and Inter-Dataset Variation) remain in datasets which dramatically lead to performance degradation. Specifically, dataset variations will lead to severe over-fitting easily and perform poor generalization in recent in-the-wild datasets which severely harm the robustness of facial landmark detection algorithm. In this study, we show that model robustness can be significantly improved by lever-aging rich variations within and between different datasets. this is non-trivial because of the serious data bias within one certain dataset and inconsistent landmark definitions between different datasets, which make it an extraordinarily tough task. To address the mentioned problems, we proposed a novel Deep Coupling Neural Network (DCNN), which consists of two strong coupling sub-networks, e.g., Dataset-Across Network (DA-Net) and Candidate-Decision Network (CD-Net). In particular, DA-Net takes advantage of different characteristics and distributions across different datasets, while CD-Net makes a final decision on candidate hypotheses given by DA-Net to leverage variations within one certain dataset. Extensive evaluations show that our approach dramatically outperforms state-of-the-art methods on the challenging 300-W and WFLW dataset (C) 2019 Elsevier Ltd. All rights reserved.

关键词： computer vision Facial landmark detection Deep learning

来源：评论

学校读者我要写书评

暂无评论

computer analysis and restoration of motion blurred images 28

Computer analysis and restoration of motion blurred images

引用

28th international conference on computer graphics and vision, graphicon 2018

作者： Melnikova, A.A. Krylov, A.S. Faculty of Computational Mathematics and Cybernartics Lomonosov Moscow State University Moscow Russia

Histogram based and correlation based algorithms for computer analysis and restoration of motion blurred images have been proposed. the histogram algorithm uses ridge detection method and direction distribution histogram analysis. the correlation algorithm finds a detection function, that when superimposed on the Fourier spectrum of the motion blurred image, gives a minimum. the results of comparison between the histogram method and the correlation method are presented. the tests show practically good results for the realistic motion blur lengths. © graphicon 2018 - 28th international conference on computer graphics and vision. All rights reserved.

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

Deep Intrinsic Image Decomposition Using Joint Parallel Learning 1

引用

36th computer graphics international conference (CGI)

作者： Yuan, Yuan Sheng, Bin Li, Ping Bi, Lei Kim, Jinman Wu, Enhua Shanghai Jiao Tong Univ Dept Comp Sci & Engn Shanghai 200240 Peoples R China Northwestern Polytech Univ Sch Marine Sci & Technol Xian 710072 Shaanxi Peoples R China Macau Univ Sci & Technol Fac Informat Technol Macau 999078 Peoples R China Univ Sydney Sch Informat Technol Biomed & Multimedia Informat Technol Res Grp Sydney NSW 2006 Australia Univ Macau Fac Sci & Technol Macau 999078 Peoples R China Chinese Acad Sci Inst Software State Key Lab Comp Sci Beijing 100190 Peoples R China

ISBN: (数字)9783030225148

ISBN: (纸本)9783030225148;9783030225131

Intrinsic image decomposition is a highly ill-posed problem in computer vision referring to extract albedo and shading from an image. In this paper, we regard it as an image-to-image translation issue and propose a novel thought, which makes use of parallel convolutional neural networks (ParCNN) to learn albedo and shading with different spatial features and data distributions, respectively. At the same time, the energy is preserved as much as possible under the constraint of image reconstruction loss shared by the two networks. Moreover, we add the gradient prior based on the traditional image formation process into the loss function, which can lead to a performance improvement of our basic learning model by jointing advantages of the physically-based method and the data-driven method. We choose MPI Sintel dataset for model training and testing. Quantitative and qualitative evaluation results outperform the state-of-the-art methods.

关键词： Intrinsic image decomposition ParCNN Gradient priors

来源：评论

学校读者我要写书评

暂无评论

16th international conference on computers Helping People with Special Needs, ICCHP 2018

16th International Conference on Computers Helping People wi...

引用

16th international conference on computers Helping People with Special Needs, ICCHP 2018

ISBN: (纸本)9783319942766

the proceedings contain 99 papers. the special focus in this conference is on computers Helping People with Special Needs. the topics include: Stereo vision based distance estimation and logo recognition for the visually impaired;intersection navigation for people with visual impairment;indoor localization using computer vision and visual-inertial odometry;hapticrein: Design and development of an interactive haptic rein for a guidance robot;echoVis: Training echolocation using binaural recordings – initial benchmark results;tactiBelt: Integrating spatial cognition and mobility theories into the design of a novel orientation and mobility assistive device for the blind;virtual navigation environment for blind and low vision people;visual shoreline detection for blind and partially sighted people;3D-printing of personalized assistive technology;camassia: Monocular interactive mobile way sonification;a proposed method for producing embossed dots graphics with a 3D printer;accessibility as prerequisite for the production of individualized aids through inclusive maker spaces;Hackability: A methodology to encourage the development of DIY assistive devices;Universal design tactile graphics production system BPLOT4 for blind teachers and blind staffs to produce tactile graphics and ink print graphics of high quality;a user study to evaluate tactile charts with blind and visually impaired people;concept-building in blind readers with thematic tactile volumes;augmented reality for people with visual impairments: Designing and creating audio-tactile content from existing objects;recording of fingertip position on tactile picture by the visually impaired and analysis of tactile information;designing an interactive tactile relief of the meissen table fountain;one-handed braille in the air.

关键词：

来源：评论

学校读者我要写书评

暂无评论

16th international conference on computers Helping People with Special Needs, ICCHP 2018

16th International Conference on Computers Helping People wi...

引用

16th international conference on computers Helping People with Special Needs, ICCHP 2018

ISBN: (纸本)9783319942735

the proceedings contain 179 papers. the special focus in this conference is on computers Helping People with Special Needs. the topics include: Stereo vision based distance estimation and logo recognition for the visually impaired;intersection navigation for people with visual impairment;indoor localization using computer vision and visual-inertial odometry;hapticrein: Design and development of an interactive haptic rein for a guidance robot;echoVis: Training echolocation using binaural recordings – initial benchmark results;tactiBelt: Integrating spatial cognition and mobility theories into the design of a novel orientation and mobility assistive device for the blind;virtual navigation environment for blind and low vision people;visual shoreline detection for blind and partially sighted people;3D-printing of personalized assistive technology;camassia: Monocular interactive mobile way sonification;a proposed method for producing embossed dots graphics with a 3D printer;accessibility as prerequisite for the production of individualized aids through inclusive maker spaces;Hackability: A methodology to encourage the development of DIY assistive devices;Universal design tactile graphics production system BPLOT4 for blind teachers and blind staffs to produce tactile graphics and ink print graphics of high quality;a user study to evaluate tactile charts with blind and visually impaired people;concept-building in blind readers with thematic tactile volumes;augmented reality for people with visual impairments: Designing and creating audio-tactile content from existing objects;recording of fingertip position on tactile picture by the visually impaired and analysis of tactile information;designing an interactive tactile relief of the meissen table fountain;one-handed braille in the air.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Human image deep binary descriptor for person reidentification and tracking in video 27

Human image deep binary descriptor for person reidentificati...

引用

27th international conference on computer graphics and vision, graphicon 2017

作者： Liutov, V.S. Konushin, A.S. MSU Moscow Russia HSE Moscow Russia

ISBN: (纸本)9785794429633

the work focuses on human reidentification, i.e. identifying an unknown person using a photo from a surveillance camera. A base method that involves modifying the VGG16 neural network algorithm is proposed. Experimental comparison of the algorithm proved to be highly accurate compared to similar approaches. Furthermore, an algorithm modification, which creates binary visual descriptors that have comparable accuracy with the original descriptor, is presented. © graphicon 2017 - computer graphics and *** right reserved.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Photo-Realistic Simulation of Road Scene for Data-Driven Methods in Bad Weather 16

Photo-Realistic Simulation of Road Scene for Data-Driven Met...

引用

16th IEEE international conference on computer vision (ICCV)

作者： Li, Kunming Li, Yu You, Shaodi Barnes, Nick CSIRO Data61 Canberra ACT Australia Australian Natl Univ Canberra ACT Australia Adv Digital Sci Ctr Singapore Singapore

ISBN: (纸本)9781538610343

Modern data-driven computer vision algorithms require a large volume, varied data for validation or evaluation. We utilize computer graphics techniques to generate a large volume foggy image dataset of road scenes with different levels of fog. We compare with other popular synthesized datasets, including data collected both from the virtual world and the real world. In addition, we benchmark recent popular dehazing methods and evaluate their performance on different datasets, which provides us an objectively comparison of their limitations and strengths. To our knowledge, this is the first foggy and hazy dataset with large volume data which can be helpful for computer vision research in the autonomous driving.

关键词： computer vision Meteorology Roads Computational modeling Autonomous vehicles Rendering (computer graphics) Image color analysis

来源：评论

学校读者我要写书评

暂无评论

PVNN: A neural network library for photometric vision 16

PVNN: A neural network library for photometric vision

引用

16th IEEE international conference on computer vision (ICCV)

作者： Yu, Ye Smith, William A. P. Univ York Dept Comp Sci York N Yorkshire England

ISBN: (纸本)9781538610343

In this paper we show how a differentiable, physics-based renderer suitable for photometric vision tasks can be implemented as layers in a deep neural network. the layers include geometric operations for representation transformations, reflectance evaluations with arbitrary numbers of light sources and statistical bidirectional reflectance distribution function (BRDF) models. We make an implementation of these layers available as a neural network library (PVNN) for theano. the layers can be incorporated into any neural network architecture, allowing parts of the photometric image formation process to be explicitly modelled in a network that is trained end to end via backpropagation. As an exemplar application, we show how to train a network with encoder-decoder architecture that learns to estimate BRDF parameters from a single image in an unsupervised manner.

关键词： Rendering (computer graphics) Neural networks Estimation Image reconstruction computer architecture Transforms

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：