← Back to team
Pamuditha Somarathne

Pamuditha Somarathne

PhD Student

I am a PhD student working at the intersection of human-centered AI and machine learning, with a focus on modeling and understanding human motion. My research explores representation learning methods for predicting, synthesizing, and analyzing human movement from multimodal and biological signals, aiming to enable accurate pose estimation and robust motion understanding in realistic, interactive settings. I am also interested in biological signal processing, including applications of PPG and PCG for health monitoring.

Research Interests

Human Motion ModellingBiological Signal Processing

Publications

Journal / IEEE T-BME / 2026

A Dual Classifier-Regressor Architecture for Heart Sound Onset/Offset Detection

P Somarathne, S Herath, G Gargiulo, P Breen, N Anderson, Y Yao, T Liu, A Withana

Conference / AHs '25 / 2025

Just Before Touch: Manipulating Perceived Haptic Sensations through Proactive Vibrotactile Cues in Virtual Reality

Y Dong, P Somarathne, CT Jin, J Kim, A Bianchi, A Withana

Conference / OzCHI '25 / 2025

NeverLagging: Enhancing Virtual Reality Finger Tracking with a Physics-Inspired Time-Agnostic Graph Neural Network

T Li, P Somarathne, Z Sarsenbayeva, A Withana

DOI |
Preprint / arXiv / 2025

TA-GNN: Physics Inspired Time-Agnostic Graph Neural Network for Finger Motion Prediction

T Li, P Somarathne, Z Sarsenbayeva, A Withana

Conference / ISWC '24 / 2024

Efficient and Robust Heart Rate Estimation Approach for Noisy Wearable PPG Sensors Using Ideal Representation Learning

A Niwarthana, P Somarathne, P Qian, KT Yong, A Withana

Conference / AHs '24 / 2024

PairPlayVR: Shared Hand Control for Virtual Games

H Zhou, P Somarathne, TA Peirispulle, C Fan, Z Sarsenbayeva, ...

aid-lab

School of Computer Science

The University of Sydney

1 Cleveland St, Darlington NSW 2008, Australia

Contact

© 2026 aid-lab. All rights reserved.