Harshana Weligampola

I am a Ph.D. student working as a research assistant at Purdue University supervised by Prof. Stanley Chan, where I am developing solutions to projects related to computational imaging.

I graduated from the University of Peradeniya in 2020 with a B.Sc specializing in Computer Engineering. Apart from research, I am also interested in competitive programming, and I like to listen to music.

Email  /  CV  /  Google Scholar  /  Github

profile photo
Research

I'm interested in computer vision, machine learning, optimization, and image processing.

iid An Optical Physics inspired CNN approach for Intrinsic Image Decomposition
Harshana Weligampola, Gihan Jayatilaka, Suren Sritharan, Parakrama Ekanayake, Roshan Ragel, Vijitha Herath Roshan Godaliyadda,
IEEE International Conference on Image Processing (ICIP), 2021
preprint

A CNN model to solve the IID problem trained in an unspervised manner using physics-based parameters derived from an image.

ga_b5g Use of Artificial Intelligence on spatio-temporal data to generate insights during COVID-19 pandemic: A Review
Gihan Jayatilaka, Jameel Hassan, Umar Marikkar, Rumali Perera, Suren Sritharan, Harshana Weligampola, Mevan Ekanayake, Roshan Godaliyadda, Parakrama Ekanayake, Vijitha Herath, Dilshan Godaliyadda, Anuruddhika Rathnayake, Samath D. Dharmaratne, Janaka Ekanayake
medRxiv, 2020
preprint

Review of the use of AI techniques for spatio-temporal modeling, forecasting, and impact modeling on diverse populations as it relates to COVID-19 and its potential for future applications.

ga_b5g A Study on Deep Learning for Latency Constraint Applications in Beyond 5G Wireless Systems
Suren Sritharan, Harshana Weligampola, Haris Gačanin
IEEE Access, 2020
preprint

Study of the practical limitations of learning models in a dynamic wireless environment.

A Retinex Based GAN Pipeline to Utilize Paired and Unpaired Datasets for Enhancing Low Light Images
Harshana Weligampola, Gihan Jayatilaka, Suren Sritharan, Roshan Godaliyadda, Parakrama Ekanayake, Roshan Ragel, Vijitha Herath
Moratuwa Engineering Research Conference (MERCon), 2020
preprint

Low light image enhancement utilizing paired and unpaired dataset using cyclic consistency.

Generalizing Foreground Estimation Algorithms in Dynamic Background Conditions [ABSTRACT]
Gihan Jayatilaka*, Suren Sritharan*, Harshana Weligampola*, Dhammika Elkaduwe, Roshan Godaliyadda, Parakrama Ekanayake, Vijitha Herath Nalin Harishchandra
Sri Lanka Technological Campus International Research Conference (SLTC irc), 2020
abstract / github

Foreground estimation through background cancellation in dynamic background conditions based on classical signal processing techniques in spatio-temporal domain.

Non-contact Infant Sleep Apnea Detection
Gihan Jayatilaka*, Harshana Weligampola*, Suren Sritharan*, Pankayaraj Pathmanathan, Roshan Ragel, Isuru Nawinne
IEEE International Conference on Industrial and Information Systems (ICIIS), 2019
preprint / github

Sleep apnea identification in infants through a video feed.

Teaching Experience
SLTC Teaching assistant @ SLTC.
Programming fundamentals (ECS100), 2020.
Internet Technologies (CCS112), 2020.
Communication Protocols (CCS201), 2021.
UOP Volunteer Instructor @ UOP
Computer vision (COxxx), 2020.
Software architecture (COxxx), 2020.

Credits to Jon Barron.
Last updated June 2021.