SAAD ALI
5939 5th Ave, Apt 22, Pittsburgh, PA 15232, USA
Phone: 321-276-3314, Email: saada@cs.cmu.edu, Website http://www.cs.ucf.edu/~sali
EDUCATION
2008-to date: Postdoctoral Fellow, The Robotics Institute, Carnegie Mellon University, USA
RESEARCH ADVISOR: Dr.Takeo Kanade
2004–2008: Doctorate of Philosophy in Computer Science, University of Central Florida, USA
RESEARCH AREA: Computer Vision
RESEARCH ADVISOR: Dr. Mubarak Shah, Agere Chair Professor, University of Central Florida
DISSERTATION: Taming Crowded Visual Scenes
1999–2003: Bachelor of Science in Computer Engineering, Ghulam Ishaq Khan Institute, Pakistan
BACHELOR THESIS: Developing Collective Intelligence in a Colony of Robots Using Neuro-Evolution
RESEARCH INTERESTS
Various areas of my interests in the field of Computer Vision are:
-
Surveillance, with a focus on high density crowded scenes.
| -
Object & Scene Recognition
| -
Human Action Recognition & Synthesis
| | | | |
RESEARCH PUBLICATIONS
Submitted
| -
Saad Ali, Vladimir Reilly and Mubarak Shah, Motion and Appearance Contexts for Tracking and Re-Acquiring Targets in Aerial Videos, IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI).
-
Saad Ali & Mubarak Shah, Kinematic Features based Action Recognition Framework, IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI).
-
Saad Ali & Mubarak Shah, Context Coherency for Object Categorization, Journal of Computer Vision and Image Understanding (CVIU).
|
Major Conferences
| -
Saad Ali, & Mubarak Shah, Floor Fields for Tracking in High Density Crowded Scenes, The 10th European Conference on Computer Vision (ECCV), 2008. (Oral Presentation).
-
Jingen Liu, Saad Ali, & Mubarak Shah, Recognizing Human Actions Using Multiple Features, IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2008.
-
Saad Ali, Arslan Basharat & Mubarak Shah, Chaotic Invariants for Human Action Recognition, IEEE International Conference on Computer Vision (ICCV), 2007.
-
Paul Scovanner, Saad Ali & Mubarak Shah, A 3-Dimensional SIFT Descriptor and its Application to Action Recognition, ACM Multimedia (ACM MM), 2007.
-
Saad Ali & Mubarak Shah, A Lagrangian Particle Dynamics Approach for Crowd Flow Segmentation and Stability Analysis, IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2007. (Oral Presentation).
-
Saad Ali, Vladimir Reilly & Mubarak Shah, Motion and Appearance Contexts for Tracking and Re-Acquiring Targets in Aerial Videos, IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2007.
-
Saad Ali & Mubarak Shah, A Supervised Learning Framework for Generic Object Detection in Images, IEEE International Conference on Computer Vision (ICCV), 2005.
-
Omar Javed, Saad Ali & Mubarak Shah, Online Detection and Classification of Moving Objects Using Progressively Improving Detectors, IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2005.
-
Saad Ali & Mubarak Shah, An Integrated Approach for Generic Object Detection Using Kernel PCA and Boosting, IEEE International Conference on Multimedia and Expo (ICME), 2005.
|
Workshops
| -
Saad Ali & Mubarak Shah, COCOA - Tracking in Aerial Imagery, SPIE Airborne Intelligence, Surveillance, Reconnaissance (ISR) Systems and Applications, 2006.
-
Yun Zhai, Jingen Liu, Xiaochun Cao, Arslan Basharat, Asaad Hakeem, Saad Ali, Mubarak Shah, Costantino Grana & Rita Cucchiarra, Video Understanding and Content-Based Retrieval, TREC Video Retrieval Evaluation Forum (TRECVID), Gaithersburg, 2005.
-
Yun Zhai, Xiaochun Cao, Yunjun Zhang, Omar Javed, Alper Yilmaz, Fahd Rafi, Saad Ali, Orkun Alatas, Saad Khan, & Mubarak Shah, University of Central Florida at TRECVID 2004, TREC Video Retrieval Evaluation Forum (TRECVID), Gaithersburg, 2005.
|
Talks
-
“Crowd Flow Segmentation and Instability Analysis”, IEEE CVPR, Minneapolis, MN, 2007.
-
“Visual Analysis of Crowded Scenes”, Department of Bioimaging Sciences, Yale University, New Haven, CT, 2008.
-
“Visual Analysis of Crowded Scenes”, Carnegie Mellon University Robotics Institute VASC Seminar Series, Pittsburgh, PA, 2008.
-
“Segmentation and Tracking in Crowded Scenes”, Department of Statistics Speakers Series, University of California, Los Angeles, CA, 2008.
-
“Visual Analysis of Crowded Scenes”, Department of Computer Science, University of North Carolina, Chapel Hill, NC, 2008.
Demos
-
“COCOA – Tracking in Aerial Imagery” in IEEE ICCV 2005: Developed a complete system, called COCOA, capable of performing tasks such as motion compensation, moving object detection, object segmentation, object tracking and indexing of videos taken from a camera mounted on a moving aerial platform (e.g. UAVs). For each task, several algorithms were integrated into the system. The first working version of the system was coded in Matlab which is now converted into JAVA for faster processing and platform independence.
Website: http://www.cs.ucf.edu/~vision/projects/COCOAWebsite/CocoaWebsite/featured_project.html
RESEARCH PROJECTS
-
Night Time Surveillance, Perceptek (DARPA’s STTR)
-
Co-Training for Online Learning: Developed a co-training based approach to continuously label incoming data, and also employed it for online update of the boosted classifier. Website: http://www.cs.ucf.edu/~sali/Projects/CoTrain/index.html
-
Human Classification: Worked on developing a boosting based human classifier for night-time imagery using features derived from Kernel PCA. Website: http://server.cs.ucf.edu/~vision/projects/ICCVPaperWebsite/featured_project.html
-
VACE Phase II, Disruptive Technology Office
-
Detection and Tracking in UAV Imagery: Handled the core-evaluation of COCOA system on task of detection and tracking of moving targets in UAV imagery.
Website: http://www.cs.ucf.edu/~vision/projects/COCOAWebsite/CocoaWebsite/featured_project.html
-
High Definition Aerial Video Analysis: Analyzed the pros and cons of using high definition aerial videos for ground target detection and tracking.
-
VACE Phase III, Disruptive Technology Office
-
Crowd Flow Segmentation: Developed an algorithm that segments a crowded scene into different crowd entities by integrating motion information through Lagrangian Particle dynamics.
Website: http://www.cs.ucf.edu/~sali/Projects/CrowdSegmentation/index.html
-
Chaotic Invariants for Action Recognition: Developed an algorithm that used theory of chaotic systems for human action modeling and recognition in videos.
Website: http://www.cs.ucf.edu/~sali/Projects/ChaoticInvariants/index.html
-
Context Coherency for Object Categorization: Developed a constraint of context coherency for discovering semantically meaningful patch grouping, where discovered groupings are then used for improved object detection.
-
Features Fusion through Fiedler Embedding: Developed a principled framework for fusing information from complementary features (patches, contours, segments) for object recognition.
-
Motion and Appearance Contexts for Target Reacquisition: Developed novel concepts of motion and appearance context for re-acquiring targets in aerial videos.
-
Aerial Video Indexing and Retrieval, Lockheed Martin, 2005
In this project we worked on developing a complete system for Event based Indexing and Retrieval of videos taken by a UAV.
Website: http://www.cs.ucf.edu/~vision/projects/COCOAWebsite/CocoaWebsite/featured_project.html
-
Exploring New Methods for Video Indexing of Aerial Videos, Lockheed Martin, 2006
In this project we worked on developing enhanced algorithms capable of performing event detection in aerial videos despite misdetections and broken trajectories. Use of multiple UAV for improved event understanding was also explored.
-
Advanced Video Analysis and Fusion, Lockheed Martin, 2007
In this project we worked on developing a framework for object association across ground surveillance and airborne cameras.
In this project we worked on developing a proof of concept for vision based missile guidance control by computing the current GPS location of the missile using images.
Website: http://www.cs.ucf.edu/~sali/Projects/ARO/index.html
-
TREC Video Retrieval Evaluation Forum (Trecvid 2004, 2005), NIST
I worked as an active member of the UCF Trecvid team during 2004 and 2005, on tasks such as automatic segmentation, indexing, and content-based retrieval of digital video.
WORK EXPEIRENCE
2005–present: Graduate Research Assistant (Dr. Mubarak Shah), University of Central Florida.
2004 – 2005: Graduate Teaching Assistant, University of Central Florida.
-
Object Oriented Programming (Lab instructor)
-
C Programming (Lab instructor for class of 50-75 students)
2003 – 2004: Software Engineer/Makabu Pvt. Ltd, Pakistan.
Honors and Professional Activities
-
Reviewer: IEEE Transactions on Pattern Analysis and Machine Intelligence, IEEE Transactions on Circuits and Systems for Video Technology, Journal of Machine Vision and Applications (MVA).
-
Best Undergraduate Thesis Award.
RESEARCH GRANT WRITING EXPERIENCE
-
Authored or co-authored following research grant proposals: SBIR/STTR (Efficient and Robust Algorithms for Real-time Tracking of Multiple Moving Targets), VACE Phase III (DTO), SAIC Image Registration, Florida Imaging Center (UCF), NSF (Analysis of Crowded Visual Scenes).
PREFFERD PLATFORMS
-
Programming/Scripting Languages and Libraries : C/C++, Matlab. Operating System : Microsoft Windows.
REFERENCES:
-
Mubarak Shah, Agere Chair Professor
Director, Computer Vision Laboratory, University of Central Florida.
shah@eecs.ucf.edu
University of Central Florida,
mtappen@eecs.ucf.edu
-
David Fleet, Professor,
Department of Computer Science, University of Toronto
fleet@cs.toronto.edu
Share with your friends: |
