The Temporal Intelligence and Motion Extraction (TIME) Lab is a unique research hub at The Australian National University (ANU) dedicated to advancing video processing, dynamics, and machine learning. There, a team of young researchers are undertaking pioneering projects and pushing the boundaries of technology to address real-world challenges.
Established in February 2024, the TIME Lab research team of honours and master’s students are led by Dr Lei Wang from the ANU School of Computing.
It is supported by strong collaborations with top researchers from Data61/CSIRO, the University of Central Florida (UCF), and Curtin University, as well as industry partners like Space Zero Investments Pty Ltd. These connections enable students to engage in meaningful projects that address real-world problems – bridging academic rigor with real-world impact.
Students at TIME Lab are driven by a shared passion for discovery. The Lab fosters a close-kin community with a collaborative and supportive environment where young researchers can be curious, innovate, and push the boundaries of computer science.
“My time at TIME Lab was incredibly rewarding. It’s inspiring to work alongside such talented and motivated individuals,” shares Gennie Nguyen (Master of Computing (Advanced) student).
Mentoring rising stars
At the helm of TIME Lab is Dr Lei Wang, a Research Fellow at the ANU School of Computing. Dr Wang drives cutting-edge research and is renowned for his expertise in computer vision and machine learning. He is deeply committed to mentoring and supporting students to grow personally and professionally.
“As a supervisor, I am honored to mentor these rising stars, who are dedicated to addressing critical challenges and driven by a vision for a better tomorrow.”
“It brings me immense satisfaction to guide, support, and inspire them as we collectively advance the frontiers of scientific knowledge, one step at a time. ANU is renowned for cultivating exceptional talent, and our students truly represent the brightest minds in science, shaping the future of innovation and discovery,” said Dr Wang.
“Working at TIME Lab has been transformative. The support and camaraderie here inspire me to push the limits of what’s possible in video processing.” – Qixiang Chen (Bachelor of Advanced Computing (Honours) student).
What has TIME Lab been up to?
Since its inception, TIME Lab has celebrated remarkable milestones and is quickly gaining recognition for its commitment to excellence and impactful contributions.
The Lab has been recognised at leading conferences, including the International Conference on Machine Learning (ICML), the Asian Conference on Machine Learning (ACML), and Conference on Neural Information Processing Systems (NeurIPS).
Final-year undergraduate student Qixiang Chen has made strides with his research project, “Motion Meets Attention: Video Motion Prompts,” presented at ACML 2024. Co-authored with Dr Lei Wang, Professor Piotr Koniusz, and Professor Tom Gedeon, Qixiang’s work introduces a novel approach to enhancing fine-grained motion recognition by integrating video motion prompts. This innovative method combines traditional video processing techniques with advanced, learnable motion prompts to enhance video understanding tasks, enabling more accurate and efficient video analysis. The proposed approach is particularly effective for fine-grained recognition tasks, such as detailed sports analysis (e.g., gymnastics, diving) or cooking.
This method could help optimise performance analysis in competitive sports, improve training feedback systems, and enhance content indexing in cooking tutorials, where recognising subtle, precise actions is crucial.
“My goal is to leverage advanced visual understanding technologies to enhance and improve experiences for people around the world. I hope to contribute to innovations that make a meaningful impact on how we interact with and understand visual content,” says Mr Chen.
Masters student Liyun Zhu’s research project, “Advancing Video Anomaly Detection: A Concise Review and a New Dataset,” was presented at the NeurIPS 2024. Collaborating with Dr Lei Wang, Arjun Raj, Professor Tom Gedeon, and Professor Chen Chen, they reviewed video anomaly detection models and datasets from various perspectives, identifying key practical challenges in real-world applications, such as the lack of a comprehensive dataset including diverse and challenging scenarios. To address this gap, they introduce the Multi-Scenario Anomaly Detection (MSAD) dataset, which is the first to feature diverse motion patterns and challenging variations, including varying lighting and weather conditions. This dataset provides a robust foundation for training more effective models to tackle real-world anomaly detection challenges.
“My passion for machine learning and computer vision drives me to continuously push the boundaries of video understanding one step at a time. I like to draw inspiration from the creativity I find in art and poetry,” says Mr Zhu.
Second-year undergraduate student Xiuyuan (Jack) Yuan recently published a paper in a recent machine learning conference. As a co-first author, Jack collaborated with Dr Lei Wang to explore an innovative use of Taylor series, a 300-year-old mathematical tool, to highlight moving objects in videos. This approach helps to classify videos into predefined human action classes, such as jogging, cooking, and riding.
By enabling efficient data extraction and privacy-preserving techniques, the proposed Taylor format reduces the risk of sensitive information exposure, benefiting applications like surveillance, healthcare, and mobile devices. Its design supports lower storage costs, making video usage more secure and accessible across industries.
TIME Lab students have been recognised with summer scholarships and industry challenge awards, further enhancing their research capabilities and professional growth. Arjun Raj, Liyun Zhu, Liwen Luo, and Qixiang Chen were awarded the Active Intelligence Corp’s 2023 Active Intelligence Research Challenge Award and Xiuyuan Yuan was supported with a Summer Research Internship at the ANU School of Computing.
Looking ahead
TIME Lab’s dedication to supporting young researchers ensures it continues to nurture the next generation of innovators. It is a growing community of early career researchers, providing opportunities for discovery from undergraduate and postgraduate students from the ANU School of Computing.
“Being at TIME Lab allows me to find my passion and explore the exciting possibilities of multimodal fusion under a supportive environment. It’s a place that fuels creativity and growth.” - Dexuan Ding (Bachelor of Advanced Computing (Honours) student)
As TIME Lab continues to advance its research and expand collaborations, the team remains at the forefront of technological innovation and scientific discovery.
“After having previously collaborated with Dr Lei Wang in the private sector, I’m excited about the potential for the TIME Lab to help bring innovation and insights to the Australian industry,” - Graeme Woods (Head of Data Analytics, FinClear)
“The TIME team has been amazing to deal with, inspiring to collaborate with, and I am excited to explore further opportunities with them. I was extremely impressed with the professionalism shown by the team. Their willingness to take on feedback and incorporate it into our projects was refreshing. Some of the most brilliant people that I have ever had the privilege to work with.” - Francis Williams (Director, Space Zero Investments Pty Ltd)
TIME Lab is committed to making a steady impact, one step at a time, in the fields of video processing and machine learning. We can’t wait to see what they produce next!
Learn more about TIME Lab: time.anu.edu.au
Learn more about studying a Bachelor of Advanced Computing (Honours) and the Masters of Computing (Advanced) at ANU