1457 Chapters
Medium 9781601323699

Untitled

Kevin Daimi, Hamid Arabnia, CSREA Press PDF

Author Index

Abdelazim, Mohamed - 244

Abdelbaki, Nashwa - 237

Abdulhadi, Mohammed M. - 367

AbManan, Jamalul-lail - 108

Airofarulla, Giuseppe - 159

Al-Bahadili, Hussein - 338

Al-Saidi, Nadia M. G. - 367

Alharthi, Raed - 274 , 280

Alshahrani , Manal - 115

Alshahrani, Hani - 274 , 280

Alshehri, Ali - 274 , 280

Alwishah, Yakeen - 54

Alzahrani, Abdulrahman - 280

Amsaad, Fathi - 206

Baldoni, Roberto - 168

Barrere, Francois - 89

Barrette, Bryce - 177

Bassous, Richard - 54

Benzekri, Abdelmalek - 89

Beyerer, Jurgen - 185

Bier, Christoph - 185

Binner, Jane - 192

Birnstill, Pascal - 185

BoBelmann, Steve - 145

Carelli, Alberto - 138 , 152

Chan, Chien-Chung - 286

Chan, Kevin - 223

Chen, Lei - 103

Chen, Yalin - 61

Chen, Yin-Jun - 47

Cheng, Sheng-Tzong - 47

Choi Ortiz, Hen Su - 3

Chou, Jue-Sam - 61

Chowdhury, Md Nasimuzzaman - 30

See All Chapters
Medium 9781601323262

An SIMD Architecture for Shortest-Path Search and Its FPGA Implementation

Hamid R. Arabnia, Lou D'Alotto, Hiroshi Ishii, Minoru Ito, Kazuki Joe, Hiroaki Nishikawa, Georgios Sirakoulis, William Spataro, Giuseppe A. Trunfio, George A. Gravvanis, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti CSREA Press PDF

Int'l Conf. Par. and Dist. Proc. Tech. and Appl. | PDPTA'14 |

53

An SIMD Architecture for Shortest-Path Search and Its FPGA

Implementation

Yasuhiro Takei, Masanori Hariyama and Michitaka Kameyama

Graduate School of Information Sciences, Tohoku University

Aoba 6-6-05, Aramaki, Aoba, Sendai, Miyagi, 980-8579, Japan

Email: {takei, hariyama, kameyama}@ecei.tohoku.ac.jp

Abstract— Shortest-path search over graphs plays an important role in various applications. However, shortest path algorithms such as the Dijkstra’s algorithm include complex processings. It is difficult for accelerators with fixed-datapath such as GPUs to accelerate these algorithms efficiently. This paper presents an

FPGA-based accelerator with a SIMD architecture for the shortest-paths algorithm. In the proposed architecture, operations in the Dijkstra’s algorithm are done with a high degree of parallelism, and the memory usage is reduced by using a memory management scheme. According to the evaluation, the proposed architecture is able to deal with graphs with more than

See All Chapters
Medium 9781601323231

Content Management in Digital Libraries

Hamid R. Arabnia, Leonidas Deligiannidis, Ray Hashemi, George Jandieri, Ashu M. G. Solo, and Fernando G. Tinetti CSREA Press PDF

���

�� �� ���

2�� ����� �� ����

� �������" ��

3��� �"�� �� ����� ������ ��

�"� +��

�� ,���

#�������� ��� ����� �� �"�

��� ��������

� 5�������� ��� �����

� ��������

��� �

����"�� ����

����� ��� ������ ��������� �������� 1"�

��� �����

� �� �

��� �

"��� ���

������ �������� �� ���� �

�������� �����" ��� ������ �

����� ����������7 �� ���

���

� �

�"� ������

� �

��� ����

�! 1"�� 5�� ��� �����

�� ��!� � ����

����� ��� ���������� ���� ��

����� �� �"�

�������� ���� �� � �� 2���

1"� ����� �� ���������� ��

��4 �� #����

� 1�

�� �������

��

������� ��

�� �"� �����

�"� ��� ��� �� #����

� 1"��� ��

��

��

�� ����

��" �

�� �������

�3���� ����������

�����

� �

���

�� ��� �

��

���� ��� ���� �� #����

� 8

��

�� ������� �"� ���� ����� ��� #����

� 8��� ��� ���� �"�

�� ���

� ��� ����

���� ��

�� �"� ������ ��

���

�"�

�!

�������� ��

� ���� ����������� �

���

�� ��� �����

��

���� ��� ����� ���

��� �� ������� ���������

#

!

"�����������

��� ��� �������� �

�! ����������� �� ��� �� �"� ����

See All Chapters
Medium 9781601323248

Video Action Recognition Using an Optical Flow Based Representation

Hamid R. Arabnia, Leonidas Deligiannidis Joan Lu, Fernando G. Tinetti, Jane You, George Jandieri, Gerald Schaefer, and Ashu M.G. Solo CSREA Press PDF

272

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

Video Action Recognition Using an Optical

Flow Based Representation

Samet Akpınar and Ferda Nur Alpaslan

Department of Computer Engineering, Middle East Technical University, Ankara, Turkey

Abstract – In this study, a new model to the problem of video action recognition has been proposed. The model is based on temporal video representation for automatic annotation of videos. Video action recognition is a field of multimedia research enabling us to recognize the actions from a number of observations, where representation of temporal information becomes important. Visual, audio and textual features are important sources for representation.

Although textual and audio features provide high level semantics, retrieval performance using these features highly depends on the availability and richness of the resources.

Visual features such as edges, corners, interest points etc. are used for forming a more complicated feature, namely, optical flow. For developing methods to cope with video action recognition, we need temporally represented video information. For this reason, we propose a new temporal segment representation to formalize the video scenes as temporal information. The representation is fundamentally based on the optical flow vectors calculated for the frequently selected frames of the video scene. Weighted frame velocity concept is put forward for a whole video scene together with the set of optical flow vectors. The combined representation is used in the action based video segment classification. Proposed method is applied to significant data sets and the results are analyzed by comparing to the state-of-the art methods.

See All Chapters
Medium 9781601323170

A Framework of Preparing Minority Students in Computer Science for the Industrial Career

Hamid R. Arabnia Azita Bahrami, Leonidas Deligiannidis, George Jandieri, Ashu M. G. Solo, and Fernando G. Tinetti CSREA Press PDF

Int'l Conf. Frontiers in Education: CS and CE | FECS'14 |

283

A Framework of Preparing Minority Students in Computer Science for the Industrial Career

Yujian Fu

Department of Electrical Engineering & Computer Science, Alabama A&M University, Normal AL, U.S.A.

Abstract— The lack of preparedness of Computer Science graduates for a professional career is a common complaint raised by industry practitioners. The career progression of many new graduates is severely impacted due to the lack of problem solving capabilities and well rounded skills.

For example, some of the technically stronger graduates lack communication and managerial skills and vise versa.

Industry based capstone projects, incorporated as a part of an undergraduate degree, are a well accepted means of preparing students for their professional careers. This paper using a curriculum design of object oriented implementation class of graduate students at the Alabama A&M University with engaging in industry based projects at the first year of their degree to illustrate how to increase student motivation, improve software development capability, increase problemsolving capability and speed up the skills and knowledge of engineering discipline. The project developed in the curriculum aims at providing students a real-world software engineering development experience, improving student programming skills and project management capability and object oriented design concepts, and preparing them for industry. At the end of semester, we observed a successful achievement of these projects in giving aspects of evaluation. We believe this achievement is due to the practice on the object oriented programming language and project design framework. Finally, we believe this approach can be extended to any other objective problem solving framework, focusing on managerial, engineering and science classes.

See All Chapters

See All Chapters