ARSLAN
BASHARAT
Phone: (407) 823 4733, Email: arslan@cs.ucf.edu, Website: www.cs.ucf.edu/~arslan
Education
In
progress: PhD in Computer Science, University of Central
Florida, Orlando, FL,
Research
Area: Computer Vision,
Research
Advisor: Dr. Mubarak Shah, Agere Chair Professor,
2006:
MS in Computer Science,
2001:
BS in Computer System Engineering, Ghulam Ishaq Khan
Institute, Pakistan.
Research Interests
I am interested in various areas in computer vision research, including:
|
·
Dynamic
Texture Modeling & Synthesis ·
Human
Action Tracking, Recognition, & Synthesis ·
Nonlinear
Dynamical Systems & Chaos Theory |
·
3D
Scene Understanding ·
Video
Surveillance ·
Content
Based Video Matching |
Publications
·
Geometric Constraints on 2D Action Models for Tracking Human Body, Alexei Gritai, Arslan Basharat, Mubarak Shah, Accepted for oral
presentation at the International Conference on Pattern Recognition (ICPR), Dec 2008.
·
Learning Object Motion Patterns for Anomaly Detection and Improved
Object Detection, Arslan Basharat, Alexei
Gritai, Mubarak Shah, IEEE Conference on Computer Vision and Pattern
Recognition (CVPR), June 2008.
·
Content Based Video Matching Using Spatiotemporal Volumes, Arslan Basharat, Yun Zhai, Mubarak Shah, Journal of Computer
Vision and Image Understanding (CVIU),
2008.
·
Chaotic
Invariants for Human Action Recognition, Saad Ali, Arslan Basharat, Mubarak Shah, IEEE
International Conference on Computer Vision (ICCV), 2007.
·
Person and Vehicle Tracking in Surveillance Video, Andrew Miller, Arslan Basharat, Brandyn White, Jingen Liu,
Mubarak Shah, CLEAR Evaluation
Campaign and Workshop, 2007.
·
Detection
and Tracking of Objects from Multiple Airborne Cameras,
Mubarak Shah, Asaad Hakeem, Arslan Basharat, Publication of SPIE, March 2006.
·
University of Central Florida at TRECVID 2006 High-Level Feature
Extraction and Video Search, Jingen Liu, Yun Zhai,
Arslan Basharat, Bilal Orhan, Saad M. Khan, Humera Noor, Phillip Berkowitz,
Mubarak Shah, TREC Video Retrieval Evaluation Forum (TRECVID) 2006.
·
Looking
for Targets, Arslan Basharat, Asaad Hakeem, Mubarak Shah,
and Abhijit Mahalanobis, OE Magazine,
Member Publication of SPIE, November 2005.
·
Video
Understanding and Content-Based Retrieval, Yun Zhai,
Jingen Liu, Xiaochun Cao, Arslan Basharat, Asaad Hakeem, Saad Ali, Mubarak
Shah, Costantino Grana, Rita Cucchiarra, TREC Video Retrieval Evaluation Forum
(TRECVID), 2005.
·
A
Framework for Intelligent Sensor Network with Video Camera for Structural
Health Monitoring of Bridges, Arslan Basharat,
Necati Catbas, Mubarak Shah, Workshop on Sensor Networks and Systems for
Pervasive Computing (PerSeNS), in conjunction with PerCom 2005.
·
Challenges
in Structural Health Monitoring, Necati Catbas,
Mubarak Shah, Jason Burkett, Arslan Basharat, International Workshop on
Structural Control (IWSC), 2004.
Research
Projects
·
VACE Phase III, Disruptive Technology Office
o Prediction in Nonlinear
Dynamical Systems for Video Synthesis, 2008
We use concepts from chaotic systems theory to forecast time series. This
approach has been applied to texture and human action synthesis. We demonstrate
that our approach outperforms previous popular approaches for dynamic texture
synthesis.
o Scene Understanding in Video
Surveillance, 2008
[http://www.cs.ucf.edu/~arslan/surveillance]
We investigate new methods to integrate a higher-level layer to
the traditional surveillance pipeline to perform anomalous event detection and
scene model feedback. A novel framework for learning patterns of motion and
size of objects in static surveillance cameras has been developed.
o Human Body Parts Tracking using
Geometrical Constraints, 2008
We investigate the important problem of detection, and tracking body
joints during unusual actions. The tracking phase is guided by the joint
predictions that are computed through the geometric transformation of the model
action onto the test action.
o Chaotic Invariants for Human
Action Recognition, 2007
[http://www.cs.ucf.edu/~sali/Projects/ChaoticInvariants]
We have
used chaotic systems to model and analyze nonlinear dynamics of human actions,
where trajectories of body joints are used as the representation of the
nonlinear dynamical system generating the action.
·
VACE Phase II, Disruptive Technology Office
o Content Based Video Matching
Using Spatiotemporal Volumes, 2007
[http://www.cs.ucf.edu/~arslan/vidmatching]
This project focused on matching telecast videos based on a new
method of object level matching. We extract the foreground and background
spatiotemporal volumes from the video through motion segmentation. Earth
Mover's Distance (EMD) is used as cost measure between volume features and a
graphical model is used for volume correspondence.
·
ICCV 2005 Computer Vision Contest: Where Am I?
[http://www.cs.ucf.edu/~arslan/whereami.htm]
I actively participated in a team of four members. Given a set
of images taken at known GPS locations, the task was to find GPS locations of
test images taken at unknown locations. Our approach is based on triangulation
from image triplets. Fundamental matrices are computed through SIFT matches.
Then Kruppa’s equations are used to determine camera parameters. In addition,
degenerate cases are also handled. Our algorithm received an honorable mention at ICCV 2005.
·
[http://server.cs.ucf.edu/~vision/projects/COCOAWebsite/CocoaWebsite/featured_project.html]
I helped in this
project by implementing a blob tracker module for
·
Visual
Inspection of Railroad Tracks,
[http://www.cs.ucf.edu/~arslan/fdot2.htm]
Led
a team of three members for this project where state of art visual defect
detection system is being developed. A remotely controlled hi-rail vehicle has
been developed which successfully runs the inspection system under challenging
illumination and track conditions.
·
CLEAR
Evaluation Campaign, CHIL&NIST, 2007
I was responsible for person and vehicle tracking tasks in
ground surveillance videos. KNIGHT surveillance engine was modified for this
task.
·
TREC Video Retrieval
Evaluation Forum (TRECVID), NIST, 2005-6
I developed an approach for high-level feature extraction and
video retrieval. During 2006, our performance on high-level feature task is top
5 of 35 teams, and for interactive search our run was ranked top 4 of 36 total
runs from all over the world.
·
Visual
Monitoring of Railroad Grade Crossing,
[http://www.cs.ucf.edu/~arslan/fdot1.htm]
This included design
and testing of a surveillance system with self sustained solar power and
wireless communication. I successfully led a team of three members to take this
surveillance project through final stages of development, testing, and
delivery.
Honors and Professional Activities
·
Honorable
mention for ICCV 2005 Computer Vision Contest: Where Am I?
[http://research.microsoft.com/iccv2005/Contest]
·
Reviewer
for: TPAMI, TCSVT, MVA, PR Letters, CVPR, ICCV, and ICPR.
·
Mentored
Student Projects:
o 2008: Vision
Guided Urban Area Navigation for Unmanned Ground Vehicles, BS Senior Year
Design Project, GIK Institute, Pakistan.
o 2006:
Abandoned Baggage Detection in Surveillance Videos,
·
Graduate Research Fellowship, University
of Central Florida, 2006.
·
Dean’s
honor list at GIK Institute, Pakistan.
·
Silver
medal and merit scholarship from AJK Board of Intermediate Education, Pakistan.
Work Experience
2003-present:
University of Central Florida
·
Research
Assistant, Computer Vision Laboratory
·
Teaching
Assistant, Department of Computer Science
2001-2003:
Advanced Engineering Research Organization, Pakistan
·
Software
Design Engineer
o Radar Data Extractor: Worked on a real-time
application for data acquisition and analysis. This involved development in C++
and live testing of the system.
o Radar Display Scope: Designed & developed a QT GUI application
to display live radar picture including raw video and synthetic track
information.
2000:
Vroom Technologies, Pakistan
·
Summer
Intern
o Developed Oracle and Visual C++ based DBMS.
Preferred Platforms
Programming/Scripting
Languages and Libraries: C/C++, Matlab, and Java
Operating
System: Microsoft Windows, Linux, and MacOS X.
References
Available
upon request.