Teaching |
| Data Engineering |
| Informatics Institute of Technology, Sri Lanka (2004/2005) |
| [class page] |
| This class discusses advanced features of Oracle PlSql, Object Oriented Databases, Data Warehousing,
Data Mining, OLAP, and Distributed Transaction Processing |
Teaching Assistantships |
| Logic and Discrete Mathematics |
| Washington University in St. Louis (Fall, 2005) |
| [class page] |
| This class discusses Logic and Discrete mathematics, and its use in computer programming |
| |
| Database Management Systems |
| Washington University in St. Louis (Fall, 2005) |
| [class page] |
| This class discusses database management systems with a focus on Microsoft SQL Server |
Courses Taken |
Ph.D. & M.Sc. in Computer Science (May 2007)
Washington University in St. Louis |
- Advanced Algorithms
- Advanced Computer Graphics
- Computational Geometry
- Computer Graphics
- Computer Systems Architecture I
- Computer Vision [homeworks]
- Intelligent Data Analysis
-
- Introduction to Formal Languages and Automata
- Masters Thesis
- Mobile Robotics
- Programming Systems and Languages
- Research Seminar on Geometry Processing
- Research Seminar on Mathematical Modeling
|
B.Sc. in Information Systems (September 2004)
Manchester Metropolitan University, UK |
- Artificial Neural Networks
- Communication Networks
- Computer Systems Architecture
- Computer Systems Organization
- Computing Mathematics
- Data Engineering
- Database Systems
- Operating Systems
- Professional and Legal Issues
- Program and Data Structures
- Programming
- Information Systems Development
- Information Systems Management
- Information Technology and Information Systems
- Software Development
- Systems Analysis and Design
- Symbolic Artificial Intelligence
- Undergraduate Research
|
|
Class Project: Taco |
| A Dual Threaded Processor using Fine Grained Multithreading with Branch Supplantment |
| Sasakthi S. Abeysinghe, Richard Hough, Gazihan Alankus, Nathan Flory (2006) |
[report]
|
| Taco is a 5-staged pipelined processor with hardware support for two thread contexts which
during tests demonstrated a speedup of above 2.0x when running in dual-threaded mode. The key
concept behind obtaining this performance boost is the two Instruction Fetch (IF) and Instruction
Decode (ID) stages which are used to mask pipeline stalls due to data hazards and instruction cache
misses. |
Class Project: Frankenstein |
| A "Live"wire Segmentation Tool |
| Sasakthi S. Abeysinghe, Timothy Simpson (2006) |
[project page]
|
| Frankenstein is a Real-time Livewire segmentation tool designed to solve the "Segmentation" problem
in computer vision. For any given image, the image is first converted into grayscale and then run
through a standard edge detection algorithm to identify the prominent edges and their relative
strengths. Thereafter a graph is created with a node for each pixel in the image which is used
in real time to calculate the segmentation between the points that the user specifies. |
Research Project: cineVirtuale |
| Online Cinema Reservation with a Virtual Tour |
| Shriyanga Withanage, Sasakthi S. Abeysinghe (2005) |
|
| cineVirtuale is a framework for making complex 3D models available on the Internet. This is also capable of providing 3D Audio,
and Video within the model itself. User interaction is also available to a certain degree, and can be used to trigger
javascript and java methods. This was tested by applying the framework to implement online cinema reservation with
virtual walkthroughs of the theatre. |
Class Project: Gal Làlla |
| Optical Character Recognition for the Sinhalese Language |
| Sasakthi S. Abeysinghe, Buddhima Bandara, Madhuranga Lokupathirage, Manora Koththigoda (2003) |
|
| Gal Lalla is an optical character recognition solution for the Sinhala Language using a 3 Layered
Artificial Neural Network. After initial training obtained an accuracy of 96% for any given character
in the Sinhala Language. In addition NERD 2003, a general purpose tool for training any Artificial
Neural Network using a variety of training algorithms was developed. |
Research Project: Katha-Baha |
| Speech Recognition for the Sinhalese Language |
| Sasakthi S. Abeysinghe, Imalka de Silva, Tyronne de Silva, Prabath Gunawardane, Madhuranga Lokupathirage (2000) |
|
Katha Baha is a speech recognition solution for the Sinhala Language using a three-layer Artificial Neural
Network. This tool integrates seamlessly to Microsoft Office, and utilizes the Dragon Speech Engine
for feature classification.
- Merit Award, Challenge 2000 Competition.
|
|
