Dr Lim Mei Kuan

Lecturer
School of IT

lim.meikuan@monash.edu
+603 5515 9738
Room 2-4-37
ORCID

Personal statement

Mei Kuan LIM is a lecturer attached to the School of Information Technology at Monash University Malaysia. She graduated from Universiti Malaysia Sarawak, Malaysia in 2007 with a First Class Honours in Computer Science. In 2015, she received her Doctoral degree from University of Malaya, Malaysia, under the scholarship of Yayasan Khazanah. She further completed her post-doctoral studies in the Department of Artificial Intelligence in the University of Malaya until 2016. Previously, she was a researcher in MIMOS Berhad from 2004 to 2007, conducting research and development in video surveillance, in particularly intelligent video analytics solutions. She was also a visiting researcher in Kingston University, United Kingdom in 2012 and 2013.  Her research interests include swarm intelligence, data and video analytics, computer vision and machine learning. She has also served as reviewer for several conferences and journals, and as an organizing committee in several conferences such as ACPR 2015 and VCIP 2013.

Academic degrees

  • Doctor of Philosophy (Computer Science, Computer Vision) (2015), University of Malaya, Malaysia
  • Degree in Computer Science (with First Class Honours) (2007), Universiti Malaysia Sarawak

Professional affiliations

Member of International Professional Bodies

  • IEEE Computer Society

Research Interests

Her primary research focus is in applying Artificial Intelligence algorithms, in particularly utilizing the Swarm Intelligence approaches to understand and solve problems in complex systems. She is currently engaged in social media analytics, utilizing visual and textual information to analyse the underlying behaviours of social media that may in turn lead to collective intelligence.

Research Projects

Title: Identifying behaviour from social media images

This project aims to investigate if images from social media contains footprint of the mood of individuals. In particularly, this project involves crawling of social media images and classifying those using deep learning models. Ideally, this project would be able to detect and make predictions of users’ behaviour or pattern by transforming and extracting signals from dizzying amounts of publicly available images. Automated detection and analysis of social media may potentially help to identify depressed or otherwise at-risk individuals through large-scale passive monitoring of social media, and in the future may complement existing screening procedures.

Openings: 1 Master (by Research) Position (full-time and fully sponsored)

Title: Discovering Bias in Deep Learning Models using Metamorphic Testing

A key concern for the wider application of deep learning based models is their reputation as a “black box” approach, where the models are created directly from the data using machine learning algorithms, and most of the time, the predictive models are so complicated that the computer scientists that design these models would not be able to interpret how the predictions are made. Since deep learning based models are driven by data, this study aims to discover if bias occurs in these models when making predictions. Subsequently, if there is bias, this project aims to investigate the factors that could potentially lead to such bias during decision making. Our team adopts the metamorphic testing principle to discover the metamorphic relations that affects the models’ prediction, given a set of defined perturbations or attacks.

Title: A Robust Deepfake Detection Approach

Deepfake media, which present realistic AI-generated videos of people doing or saying things that are fictional can bring tremendous negative impact on how people perceive and trust public information and opinions. Deepfake may also violate human rights as they are used maliciously as a source of misinformation, manipulation and harassment. Therefore, this study aims to develop deepfake detection model that is able to identify manipulated media. With the rapid advancement of deepfake generation techniques, it is important that the devised deepfake detection model is able to evolve along with these advancements. Our team is working towards creating a deepfake solution that is robust against the different generation techniques and compression rate.

Education

Units taught

FIT1051 - Programming Fundamentals in Java

FIT9131 - Programming Foundations in Java

FIT5122 – Professional Practice

FIT3199 – Industry Work Experience

Journal

Kok, Ven Jyn; Lim, Mei Kuan; Chan, Chee Seng (2016) Crowd behavior analysis: A review where physics meets biology, Neurocomputing, (342-362), Volume: 177, Issue Number: 09252312, 10.1016/j.neucom.2015.11.021

Lim, Mei Kuan; Chan, Chee Seng; Monekosso, Dorothy; Remagnino, Paolo (2014) Refined particle swarm intelligence method for abrupt motion tracking, Information Sciences, (267-287), Volume: 283, Issue Number: 00200255, 10.1016/j.ins.2014.01.003

Lim, Mei Kuan; Tang, Szeling; Chan, Chee Seng (2014) ISurveillance: Intelligent framework for multiple events detection in surveillance videos, Expert Systems with Applications, (4704-4715), Volume: 41, Issue Number: 09574174, 10.1016/j.eswa.2014.02.003

Lim, M. K.; Chan, C. S.; Monekosso, D.; Remagnino, P. (2014) Detection of salient regions in crowded scenes, Electronics Letters, (363-365), Volume: 50, Issue Number: 00135194, 10.1049/el.2013.3993

Book

Ng, Ee Lee; Lim, Mei Kuan; Maul, Tomás; Lai, Weng Kin (2009) Investigations into particle swarm optimization for multi-class shape recognition, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), (599-606), Volume: 5507 LNCS, Issue Number: 03029743, 10.1007/978-3-642-03040-6_73

Conference

Yilmaz, Bedir; Kok, Ven Jyn; Lim, Mei Kuan; Abdullah, Siti Norul Huda Sheikh (2019) Perspective-aware loss function for crowd density estimation, Proceedings of the 16th International Conference on Machine Vision Applications, MVA 2019, 10.23919/MVA.2019.8758034

Lim, Mei Kuan; Kok, Ven Jyn; Loy, Chen Change; Chan, Chee Seng (2014) Crowd saliency detection via global similarity structure, Proceedings - International Conference on Pattern Recognition, (3957-3962), Issue Number: 10514651, 10.1109/ICPR.2014.678

Kristan, Matej; Pflugfelder, Roman; Leonardis, Aleš; Matas, Jiri; Porikli, Fatih; Čehovin, Luka; Nebehay, Georg; Fernandez, Gustavo; Vojíř, Tomáš; Gatt, Adam; Khajenezhad, Ahmad; Salahledin, Ahmed; Soltani-Farani, Ali; Zarezade, Ali; Petrosino, Alfredo; Milton, Anthony; Bozorgtabar, Behzad; Li, Bo; Chan, Chee Seng; Heng, Cherkeng; Ward, Dale; Kearney, David; Monekosso, Dorothy; Karaimer, Hakki Can; Rabiee, Hamid R.; Zhu, Jianke; Gao, Jin; Xiao, Jingjing; Zhang, Junge; Xing, Junliang; Huang, Kaiqi; Lebeda, Karel; Cao, Lijun; Maresca, Mario Edoardo; Lim, Mei Kuan; ELHelw, Mohamed; Felsberg, Michael; Remagnino, Paolo; Bowden, Richard; Goecke, Roland; Stolkin, Rustam; Lim, Samantha Yue Ying; Maher, Sara; Poullot, Sebastien; Wong, Sebastien; Satoh, Shin'Ichi; Chen, Weihua; Hu, Weiming; Zhang, Xiaoqin; Li, Yang; Niu, Zhiheng (2013) The visual object tracking VOT2013 challenge results, Proceedings of the IEEE International Conference on Computer Vision, (98-111), 10.1109/ICCVW.2013.20

Teng, Sing Loong; Chan, Chee Seng; Lim, Mei Kuan; Lai, Weng Kin (2010) Hybrid particle swarm optimisation for data clustering, Proceedings of SPIE - The International Society for Optical Engineering, Volume: 7546, Issue Number: 0277786X, 10.1117/12.852246

Tang, Sze Ling; Kadim, Zulaikha; Liang, Kim Meng; Lim, Mei Kuan (2010) Hybrid blob and particle filter tracking approach for robust object tracking, Procedia Computer Science, (2559-2567), Volume: 1, 10.1016/j.procs.2010.04.289

Kiran, Maleeha; Abdalla, Aisha Hassan; Yap, Yee Jiun; Lim, Mei Kuan (2008) A prediction module to optimize scheduling in a grid computing environment, Proceedings of the International Conference on Computer and Communication Engineering 2008, ICCCE08: Global Links for Human Development, (888-893), 10.1109/ICCCE.2008.4580733

Local Grants

Database Migration, Principal Investigator – 2021, Industry Grant, RM37, 440.

An integrated machine-learning model for predicting mental health risk using social media data, Principal Investigator - 2020-2022, Ministry of Higher Education, Fundamental Research Grant Scheme (FRGS), RM73,200.

Scientific and Technological Approach to Assess the Impact of Elevated Canopy Walkway and Bus Rapid Transit (BRT) on Traffic in Bandar Sunway, Co-Investigator - 2020-2022, Industry Grant, RM480,000

Cognitive Neural Network (CoNNet): A novel interpretable video surveillance framework for crime scene understanding based on attributes learning, Co-Investigator - 2020-2023, Ministry of Higher Education, Fundamental Research Grant Scheme (FRGS), RM94,300.

Investigation of thermo-kinetics characteristics to enhance the pyrolysis products from oil palm biomass using novel hybrid artificial intelligence model, Co-Investigator - 2019-2022, Ministry of Higher Education, Fundamental Research Grant Scheme (FRGS), RM89,400.

Human Collective Intelligence Analysis Based on Neural Networks to Support Visual Surveillance, Co-Investigator - 2019-2021, Ministry of Higher Education, Fundamental Research Grant Scheme (FRGS), RM81,200.

International Grants

Sense-ED: Developing mobile sensing and Natural Language Processing models to support the identification of, and interventional mechanisms for, eating disorder relapse, Co-Investigator - 2021, Big Idea Seed Funding Scheme, FIT, Monash University (VIC), Australia, AUD49,009 (~RM149,000).

Current supervision

Tay Heng Ee
PhD - Predicting Individual Mental Health State on Social Media via Multimodal Approach
2019 - Present
Monash University Malaysia

Seow Jia Wen (Scarlett)
PhD - Neural Network-Based Deepfake Detection
2019 - Present
Monash University Malaysia

Yasmin Mohd Zaifullizan (Co-supervision)
PhD - Investigation of thermo-kinetic characteristics to enhance the pyrolysis products from oil palm biomass using novel hybrid artificial intelligence model
2019 - Present
Monash University Malaysia

Muhammad Ikwan bin Jamaludin (Co-supervision)
PhD - Crowdsourced Data to Improve Rapid Flood Inundation Mapping and Flood Risk Assessment in Mixed Landuse Development
2020 - Present
Monash University Malaysia

Matthew Yeow (Co-supervision)
PhD - Identifying Reuse-Proneness of Object-Oriented Classes in Source Code Management Systems
2020 - Present
Monash University Malaysia

Yousra Ashfaq (Co-supervision)
MPhil - Moving Object Detection in VVC compressed Video
2020 - Present
Monash University Malaysia

Completed supervision

Ali Moltajaei Farid
Novel Approaches to Tracking Swarms of Robots
August 2018 - Present
Monash University Malaysia

Local Award/Recognition/Exhibition/Stewardship

Update Coming Soon