- Published papers in conferences such as CVPR24, ICRA24, ICCV23, etc.
- Developed datasets like PIE, JAAD, etc.
- Published research results in multiple international conferences including ICRA23, ICRA22, ICCV21, etc.
- Participated in projects researching IV23, NeurIPS22, IV22, etc.
- Proposed and maintained a database of vision-based prediction research, including papers, datasets, and metrics.
Research Experience
- Trajectory Prediction for Autonomous Driving: Focuses on multimodal trajectory generation, scene understanding, causal reasoning, and robustness study.
- Pedestrian Behavior Understanding: Includes psychological studies of pedestrians in traffic and prediction of their trajectories and actions in assistive driving settings.
- Road User Behavior Simulation: Methods for vehicles and pedestrian behavior simulation using learning-based methods and heuristic approaches parameterized based on behavioral studies.
- Benchmarking and Metrics: Approaches for data subclassing based on observed behaviors, agents' characteristics, and other contextual parameters, accompanied by novel metrics for more effective measurement of prediction and perception models' performance.
- Active Visual Search: Takes advantage of contextual information and detects areas of importance via attention mechanism to optimize object search process.
- Saliency Prediction: Studies the ability of methods for predicting saliency in images.
Education
Completed PhD in Computer Science and M.A.Sc in Computer Engineering under the supervision of Prof. John K. Tsotsos at York University in 2020 and 2015, respectively. Received B.Eng. degree in Computer Systems Engineering and B.A. in Business Management from Royal Melbourne Institute of Technology in 2010.
Background
Currently a senior staff engineer at Noah's Ark Laboratory, Canada, leading the decision-making and reasoning team, which specializes in behavior understanding, scene reasoning, and prediction. The focus of our work is on behavior prediction and generation in robotics and autonomous driving domains.
Miscellany
Maintains a database of vision-based prediction research, including related papers, datasets, and metrics.