Haowen Yu

I am a PhD student in the Department of Computer Science at the University of Maryland, College Park, advised by Cornelia Furmuller. My research primarily focuses on computer vision, with interests in event-based vision, deep learning for video analysis, and graph neural networks for structured visual data. I am particularly passionate about applying event-based vision on real-world tasks and designing efficient representation for event camera.

Before joining UMD, I earned my master of science degree in computer science from University of Massachusetts Amherst where I worked on event-based vision for human pose estimation. I have industry experience at TikTok, where I contributed to projects involving developing advertisement ranking algorithms and improving search ads revenue.

At UMD, I am part of the Perception and Robotics Group, collaborating on projects that explore the applications of event camera for various vision tasks. My work has been presented at venues such as ICRA, and I actively contribute to the open-source and academic research communities.