Dr Chang is an Associate Professor at School of Computing Technologies, RMIT University.He is an ARC Discovery Early Career Researcher Award (DECRA) Fellow between 2019-2021. After graduation, he was a Postdoc Research Associate in School of Computer Science, Carnegie Mellon University. He mainly worked on exploring multiple signals for automatic content analysis in unconstrained or surveillance videos and has achieved top performance in various international competitions. He received his Ph.D. degree from University of Technology Sydney. His research focus in this period was mainly on developing machine learning algorithms and applying them to multimedia analysis and computer vision.
Dr Xiaojun Chang is an Associate Professor in School of Computing Technologies, RMIT University, Australia. He is an ARC Discovery Early Career Researcher Award (DECRA) Fellow between 2019-2021 (awarded in 2018).
Before joining RMIT, he was a Senior Lecturer in Vision & Lanugage Group, Department of Data Science and AI, Faculty of Information Technology, Monash University from December 2018 to July 2021. He was a Postdoc Research Associate in School of Computer Science, Carnegie Mellon University after graduation, working with Prof. Alex Hauptmann. He has focused his research on exploring multiple signals (visual, acoustic, textual) for automatic content analysis in unconstrained or surveillence videos. His team has won multiple prizes from international grand challenges which hosted competitive teams from MIT, University of Maryland, Facebook AI Research (FAIR) and Baidu VIS, and aim to advance visual understanding using deep learning. For example, he won the first place in the TrecVID 2019 - Activity Extended Video (ActEV) challenge, which was held by National Institute of Standards and Technology, US.
His general research interest is to develop structured machine learning models for computer vision and multimedia tasks. He mainly investigates how to explore the information contained in videos and develop the advanced artificial intelligence systems. Recently, he focuses on the following topics, include:
- Video Analysis, including event detection, object detection, segmentation.
- Multi-Agent Reinforcement Learning.
- Vision-Language Grounding, including vision-language navigation, vision-and-dialog navigation, and medical report generation.
- Automated Machine Learning (AutoML).
While at Carnegie Mellon University, A/Prof Chang directed a project with US$600,000 in funding to use computer vision and multimedia techniques for predicting multimedia events from unconstrained videos. This two-year project successfully produced novel video prediction models and a software package, which are currently being used by the Los Alamos National Laboratory (LANL). The LANL is now better prepared to model, monitor and predict the event of interest from unconstrained video clips. This work has also been selected by the to present on the National Institute of Standards and Technology (NIST) for public presentation during the NIST Public Safety Innovation Accelerator Program.
As the team leader, A/Prof Chang led a team to develop a powerful machine learning and computer vision-based video analysis system named Event Labeling through Analytic Media Processing (E-LAMP) for the defence partner – Intelligence Advanced Research Project Activity (IARPA). The goal is to deal with different aspects of the overall problem of event detection. This system has been applied by the US defence to detect terrorist events, and has recognised by the wide community.
Collaborating with the listed company – Beijing Etrol Technologies Co. Ltd, A/Prof Chang developed a smart parking system as the leading CI. The system can find the location of the nearest empty parking spot, and direct the driver to it. The development of the smart parking system consists of the infrastructure for surveillance camera, algorithm design for vehicle detection/tracking, and a mobile app for user interface. This system has been commercialised.
A/Prof Chang is actively in the field of vision and language learning, and apply his expertise in this direction to challenging problems raised in the real-world. Collaborating with researchers from Sun Yat-sen University and The First Affiliated Hospital of Harbin Medical University, he has developed an advanced medical report generation method using computer vision and natural language processing techniques. This project will accurately read lung CT images and help diagnose COVID-19 cases. This project received media attention; it was covered by The Australian, Mirage News, ResearchNews, and AZoRobotics.
A/Prof Chang’s research has also led to inputs to mining. As the only CI, he collaborated with Woodside Energy and developed a seismic data processing system. The goal is to remove random noise in seismic data without harming useful signal. This system has been delivered together with the source code and demo to the industry partner.
- Zhang, M.,Su, S.,Pan, S.,Chang, X.,Abbasnejad, E.,Haffari, G. (2021). iDARTS: Differentiable Architecture Search with Stochastic Implicit Gradients In: Proceedings of the 38th International Conference on Machine Learning ( PMLR 2021), United States, 18 July 2021
- Ren, P.,Xiao, Y.,Chang, X.,Huang, P.,Li, Z.,Chen, X.,Wang, X. (2021). A comprehensive survey of neural architecture search: Challenges and solutions In: ACM Computing Surveys, 54, 1 - 34
- Gupta, B.,Yadav, K.,Razzak, I.,Psannis, K.,Castiglione, A.,Chang, X. (2021). A novel approach for phishing URLs detection using lexical based machine learning in a real-time environment In: Computer Communications, 175, 47 - 57
- Zhang, M.,Li, H.,Pan, S.,Chang, X.,Zhou, C.,Ge, Z.,Su, S. (2021). One-Shot Neural Architecture Search: Maximising Diversity to Overcome Catastrophic Forgetting In: IEEE Transactions on Pattern Analysis and Machine Intelligence, 43, 2921 - 2935
- Yuan, D.,Chang, X.,Huang, P.,Liu, Q.,He, Z. (2021). Self-Supervised Deep Correlation Tracking In: IEEE Transactions on Image Processing, 30, 976 - 985
- Yan, C.,Chang, X.,Luo, M.,Zheng, Q.,Zhang, X.,Li, Z.,Nie, F. (2021). Self-weighted Robust LDA for Multiclass Classification with Edge Classes In: ACM Transactions on Intelligent Systems and Technology, 12, 1 - 19
- Hu, S.,Zhu, F.,Chang, X.,Liang, X. (2021). UPDET: Universal multi-agent reinforcement learning via policy decoupling with transformers In: Proceedings of the 9th International Conference on Learning Representations (ICLR 2021), Vienna, Austria, 3-7 May 2021
- Lin, X.,Ren, P.,Xiao, Y.,Chang, X.,Hauptmann, A. (2021). Person Search Challenges and Solutions: A Survey In: Proceedings of the 30th International Joint Conference on Artificial Intelligence (IJCAI 2021), Montreal, Canada, 19-27 August 2021
- Zhang, J.,Wang, M.,Li, Q.,Wang, S.,Chang, X.,Wang, B. (2021). Quadratic Sparse Gaussian Graphical Model Estimation Method for Massive Variables In: Proceedings of the 29th International Joint Conference on Artificial Intelligence (IJCAI 2020), Yokohama, Japan, 7-15 January 2021
- Wu, M.,Pan, S.,Zhou, C.,Chang, X.,Zhu, X. (2020). Unsupervised Domain Adaptive Graph Convolutional Networks In: Proceedings of the 29th World Wide Web Conference (WWW 2020), Taipei, Taiwan, 20 - 24 April 2020