检索结果-内蒙古大学图书馆

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Valério Nogueira Hugo Oliveira José Augusto Silva Thales Vieira Krerley Oliveira Federal University of Alagoas Maceio Brazil Federal University of Alagoas (UFAL) Maceió AL Brazil

Customer behavior analysis is an essential issue for retailers, allowing for optimized store performance, enhanced customer experience, reduced operational costs, and consequently higher profitability. Nevertheless, not much attention has been given to computer vision approaches to automatically extract relevant information from images that could be of great value to retailers. In this paper, we present a low-cost deep learning approach to estimate the number of people in retail stores in real-time and to detect and visualize hot spots. For this purpose, only an inexpensive RGB camera, such as a surveillance camera, is required. To solve the people counting problem, we employ a supervised learning approach based on a Convolutional Neural Network (CNN) regression model. We also present a four channel image representation named RGBP image, composed of the conventional RGB image and an extra binary image P representing whether there is a visible person in each pixel of the image. To extract the latter information, we developed a foreground/background detection method that considers the peculiarities of people behavior in retail stores. The P image is also exploited to detect the hot spots of the store, which can later be visually analyzed. Several experiments were conducted to validate, evaluate and compare our approach using a dataset comprised of videos that were collected from a surveillance camera placed in a real shoe retail store. Results revealed that our approach is sufficiently robust to be used in real world situations and outperforms straightforward CNN approaches.

关键词： Cameras Machine learning Surveillance Histograms Real-time systems image color analysis Task analysis

来源：评论

学校读者我要写书评

暂无评论

MG-Hybrid: A Strongly Connected Components Detection Algorithm using Multiple GPUs

MG-Hybrid: A Strongly Connected Components Detection Algorit...

引用

IEEE International symposium on Circuits and Systems (ISCAS)

作者： Junteng Hou Shupeng Wang Guangjun Wu Bingnan Ma Chengxiang Si Siyu Jia Institute of Information Engineering Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China National Computer Network Emergency Response Technical Team/Coordination Center of China Beijing China

ISBN: (数字)9781728133201

ISBN: (纸本)9781728133218

Detection of strongly connected component (SCC) on the GPU has become a fundamental operation to accelerate graph computing. Existing SCC detection methods on multiple GPUs introduce massive unnecessary data transformation between multiple GPUs. In this paper, we propose a novel distributed SCC detection approach using multiple GPUs plus CPU. Our approach includes three key ideas: (1) segmentation and labeling over large-scale datasets; (2) collecting and merging the segmented SCCs; and (3) running tasks assignment over multiples GPUs and CPU. We implement our approach under a hybrid distributed architecture with multiple GPUs plus CPU. Our approach can achieve device-level optimization and can be compatible with the state-of-the-art algorithms. We conduct extensive theoretical and experimental analysis to demonstrate efficiency and accuracy of our approach. The experimental results expose that our approach can achieves 11.2×, 1.2×, 1.2× speedup for SCC detection using NVIDIA K80 compared with Tarjan's, FB-Trim, and FB-Hybrid algorithms respectively.

关键词： graphics processing units Detection algorithms image edge detection Task analysis Partitioning algorithms Parallel algorithms Electronic mail

来源：评论

学校读者我要写书评

暂无评论

E-Learning Remote Laboratory Station for computer Aided Design

E-Learning Remote Laboratory Station for Computer Aided Desi...

引用

International symposium on Electronics in Marine (ELMAR)

作者： Ingmar Besic Kerim Hodzic Almir Karabegovic Faculty ofElectrical Engineering University of Sarajevo Bosnia and Herzegovina

E-Learning solutions for computer Aided Design (CAD) require high quality graphics for positive impact and high satisfaction. Software tools used for hands-on CAD laboratory exercises depend on dedicated graphics processing Unit (GPU) to deal with complex graphics processing needed to visualize virtual models in real-time. It is challenging to offer the same level of experience to remote users as they usually cannot afford laboratory-level hardware with dedicated GPU nor such hardware can be provided to them as part of their e-Learning experience. In this paper, grouping of average CAD laboratory workstations in pairs is proposed in order to create remote stations capable of performing required CAD graphics processing for remote users while streaming it over the network for a near real-time experience. Remote station captures video and utilizes hardware HEVC encoding, as common capability of high-end dedicated GPUs, to perform low bitrate video streaming with sub-second latency. The remote station concept enables cost-effective extension of average conventional CAD laboratory to an e-Learning remote laboratory for up to 50% additional remote users in respect to the total number of the laboratory workstations.

关键词： graphics processing units Streaming media Workstations Solid modeling image coding Hardware

来源：评论

学校读者我要写书评

暂无评论

Effective Characterization of Relief Patterns

引用

computer graphics FORUM 2018年第5期37卷 83-92页

作者： Giachetti, A. Univ Verona Dept Comp Sci Verona Italy

In this paper, we address the problem of characterizing relief patterns over surface meshes independently on the underlying shape. We propose to tackle the problem by estimating local invariant features and encoding them using the Improved Fisher Vector technique, testing both features estimated on 3D meshes and local descriptors estimated on raster images created by encoding local surface properties (e.g. mean curvature) over a surface parametrization. We compare the robustness of the obtained descriptors against noise and surface bending and evaluate retrieval performances on a specific benchmark proposed in a track of the Eurographics Shape REtrieval Contest 2017. Results show that, with the proposed framework, it is possible to obtain retrieval results largely improving the state of the art and that the image-based approach is still effective when the underlying surface is heavily deformed.

关键词： Categories and Subject Descriptors (according to ACM CCS): 1.3.5 [computer graphics]: Computational Geometry and Object Modeling—Curve surface solid and object representations

来源：评论

学校读者我要写书评

暂无评论

Alignment of Local and Global Features from Multiple Layers of Convolutional Neural Network for image Classification

Alignment of Local and Global Features from Multiple Layers ...

引用

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Fernando Pereira dos Santos Moacir Antonelli Ponti University of Sao Paulo Sao Carlos Brazil

Convolutional networks have been extensively applied to obtain features spaces for classification tasks. Although those achieve high accuracy in many scenarios, typically only the top layers of the network are explored. Hence, a relevant question arises from this fact: are initial layers useful in terms of discriminative ability? In this paper, we leverage the complementary description offered by such first layers. Our method consists of features extraction in multiple layers, followed by feature selection, fusion of feature maps from the different layers, and space alignment. Through an extensive experimentation with different datasets and studying different training strategies, our results show that local information, coming from the first layers, may significantly improve the classification performance when merged with a global descriptor extracted from a top layer of the network. We report different methods for reducing the dimensionality of the local descriptors, and guidelines on how to align them so that to perform fusion. Our study encourages future studies on combining feature maps from multiple layers, which may be relevant in particular for transfer learning scenarios.

关键词： Feature extraction Principal component analysis Task analysis Training Shape image color analysis Two dimensional displays

来源：评论

学校读者我要写书评

暂无评论

Robust and Effective Method for Automatic Generation of One-Dimensional Transfer Functions

Robust and Effective Method for Automatic Generation of One-...

引用

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Rustam Mesquita Waldemar Celes Computer Science Department Pontifical Catholic University of Rio de Janeiro Brazil

In direct volume rendering, the use of transfer functions allows the visualization of specific parts of the volume, revealing significant structures such as boundaries between different materials. To interactively specify an adequate transfer function is a very challenging and time-consuming task. Besides, a user can settle with a transfer function that does not reveal the overall structure inside the volume. Hence, the importance of generating adequate transfer functions automatically. Since Kindlmann and Durkin's seminal work on the semi-automatic generation of transfer functions, many research works have been conducted attempting to improve their proposal. Despite some fine resulting visualizations, many of these works aimed to diminish the problem of boundary overlaps, in the data domain, by using two-dimensional transfer functions, thus decreasing the focus on the automatic generation and leaving to the user, again, the task to manually isolate isosurfaces. Moreover, fine-tuning two-dimensional transfer functions is less intuitive than adjusting one-dimensional ones. This paper redirects Kindlmann and Durkin's concepts of boundary characterization and proposes a new method to automatically generate one-dimensional transfer functions, more robust to boundary overlaps. This work also suggests a rendering filter to isolate detected isosurfaces in the visualization.

关键词： Transfer functions Histograms Two dimensional displays Rendering (computer graphics) Isosurfaces Proposals Standards

来源：评论

学校读者我要写书评

暂无评论

Simple and Effective Load Volume Estimation in Moving Trucks using LiDARs

Simple and Effective Load Volume Estimation in Moving Trucks...

引用

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Lucas L. Amorim Filipe Mutz Alberto F. De Souza Claudine Badue Thiago Oliveira-Santos Universidade Federal do Espírito Santo (UFES) Vitória ES Brazil Instituto Federal do Espírito Santo (IFES) Serra ES Brazil

Industries need to track the amount of materials and goods transported through processing units in order to optimize production. In large-scale industries, trucks and trains are commonly used for transportation. The manual evaluation of the volume of material being transported by these vehicles can be imprecise, inefficient, and even unsafe for employees. Therefore, this work presents an automated system for estimating the volume of load in moving trucks using a pair of multi-layer light detection and ranging (LiDAR) sensors. The sensors are mounted in a structure so that trucks can pass through without stopping. The proposed system can be used with any type of compact load such as grains, and powders. A mesh of the load is built and used for estimating the volume. A simple, efficient, and effective heuristic is proposed for tracking the truck's positions. The system was deployed and evaluated in a mining company in real conditions of operation. Experimental results indicate that the system produces accurate estimates of the volume of ore powder transported by trucks. The reconstruction of the loads and the estimative of their volumes are performed once the data is acquired and lasts less than 1.5 minutes on average.

关键词： Laser radar Three-dimensional displays Sensors Volume measurement Roads image reconstruction Load modeling

来源：评论

学校读者我要写书评

暂无评论

Generating Audiovisual Summaries from Literary Works using Emotion Analysis

Generating Audiovisual Summaries from Literary Works using E...

引用

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Daniela F. Milon-Flores Jose Ochoa-Luna Erick Gomez-Nieto Department of Computer Science Universidad Católica San Pablo Arequipa Peru Center for the Study of Violence University of São Paulo São Paulo SP Brazil

Literature work reading is an essential activity for human communication and learning. However, several relevant tasks as selection, filter or analyze in a high number of such works become complex. For dealing with this requirement, several strategies are proposed to rapidly inspect substantial amounts of text, or retrieve information previously read, exploiting graphical, textual or auditory resources. In this paper, we propose a methodology to generate audiovisual summaries by the combination of emotion-based music composition and graph-based animation. We applied natural language processing algorithms for extracting emotions and characters involved in literary work. Then, we use the extracted information to compose a musical piece to accompany the visual narration of the story aiming to convey the extracted emotion. For that, we set important musical features as chord progressions, tempo, scale, and octaves, and we assign a set of suitable instruments. Moreover, we animate a graph to sum up the dialogues between the characters in the literary work. Finally, to assess the quality of our methodology, we conducted two user studies that reveal that our proposal provides a high level of understanding over the content of the literary work besides bringing a pleasant experience to the user.

关键词： Music Visualization Feature extraction Instruments Proposals Task analysis Animation

来源：评论

学校读者我要写书评

暂无评论

Fast and Smart Segmentation of Paraspinal Muscles in Magnetic Resonance Imaging with CleverSeg

Fast and Smart Segmentation of Paraspinal Muscles in Magneti...

引用

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Jonathan S. Ramos Mirela T. Cazzolato Bruno S. Faiçal Oscar A. C. Linares Marcello H. Nogueira-Barbosa Caetano Traina Agma J. M. Traina Institute of Mathematics and Computer Science (ICMC) University of São Paulo (USP) São Carlos SP BR University of Sao Paulo Sao Carlos Brazil Ribeirao Preto Medical School (FMRP) University of São Paulo (USP) Ribeirão Preto SP BR

Magnetic Resonance Imaging (MRI) is a non-invasive technique, which has been employed to detect and diagnose many spine pathologies. In a computer-Aided Diagnosis(CAD) context, the segmentation of the paraspinal musculature from MRI may support measurement, quantification, and analysis of muscle-related pathologies. Current semi-automatic seg-mentation techniques require too much time from the physicians to annotate all slices in the exams. In this work, we focus on minimizing the time spent on manual annotation as well as on the overall segmentation processing time. We use the mean absolute error between slices aiming at minimizing the number of annotated slices in each exam. Moreover, we optimize the manual annotation time by estimating the inside annotation based on the outside annotation, while the competitors demand the annotation of inside and outside annotation (seeds). The experimental evaluation shows that our proposed approaches able to speed up the manual annotation process in up to 50%by annotating only a few representative slices, without loss of accuracy. By annotating only the outside region, the process can be further speed up by another 50%, reducing the total time to only 25% of the previously required. Thus, the total time spent on manual annotation is reduced by up to 75%, and, since human interaction is greatly diminished, allows a more productive and less tiresome activity. Despite that, our proposed CleverSeg method presented accuracy similar to or better than the competitors, while managing a faster processing time.

关键词： image segmentation Muscles Magnetic resonance imaging Three-dimensional displays Manuals Silicon Two dimensional displays

来源：评论

学校读者我要写书评

暂无评论

CAD System for Breast US images with Speckle Noise Reduction and Bio-Inspired Segmentation

CAD System for Breast US Images with Speckle Noise Reduction...

引用

brazilian symposium on computer graphics and image processing (SIBGRAPI)

作者： Paulo S. S. Rodrigues Guilherme A. Wachs Lopes Gilson A. Giraldi Celia A. Z. Barcelos Luciana Vieira Denise Guliato Bikesh Kumar Singh Centro Universitario FEI Sao Bernardo do Campo Brazil COMAC National Laboratory for Scientific Computing Petrópolis RJ Brazil Federal University of Uberlandia Uberlandia Brazil Federal University of Uberlandia FMC Uberlandia Minas Gerais Brazil National Institute of Technology Raipur Raipur India

Ultrasound (US) images are highly susceptible to speckle-like noise which makes imperative to use specific techniques for image smoothing. However, this process can lead to undesirable side effects such as the degradation of the real contour of the region of interest (ROI). In such context, this paper presents a new methodology for computer aided diagnosis (CAD) systems whose heart is the combination of a method for speckle noise reduction, with histogram equalization and a technique for image segmentation that uses the bio-inspired firefly algorithm and Bayesian model. The segmentation approach and the equalization are applied in two distinct stages: globally and locally. The global application produces an initial coarse estimate of the ROI, and the local application defines this region more precisely. In the classification step we carried out experiments which show that the combination of features computed both within and below the lesion strongly influences the final accuracy. We show that the gray-scale distribution and statistical moments within the lesion together with gray-scale distribution and contrast of the region below the lesion is the combination that produces the better classification results. Experiments in a database of 250 US images of breast anomalies (100 benign and 150 malignant) show that the proposed methodology reaches performance of 95%.

关键词： Feature extraction image segmentation Lesions Histograms Breast Speckle Noise reduction

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：