Pengyu Zhang 张鹏宇

I am a postdoctoral fellow in the Vision and Learning Lab at the University of Alberta, working with Prof. Li Cheng. Previously, I was a research fellow in the Material Robotics Lab at the National University of Singapore, supervised by Prof. Shuzhi Sam Ge.

I received my Ph.D. from Dalian University of Technology, where I was advised by Prof. Huchuan Lu. My research focuses on video perception and bio-inspired vision, with particular interests in visual tracking and segmentation, sign language understanding, and event-based vision.

Portrait of Pengyu Zhang

Research

My goal is to build visual perception systems that remain reliable under motion, modality changes, and real-world uncertainty. I work across three connected themes: multi-modal object tracking, video object segmentation, and human-centered visual understanding from signs and events.

News

The Vision and Learning Lab at the University of Alberta is recruiting graduate students and postdoctoral researchers.

One paper on video object segmentation was accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence.

We organized the first workshop and competition on Multi-Modal UAV Vision at RISEx 2025.

I received the 2024 Liaoning Provincial Outstanding Doctoral Dissertation Award.

Our workshop on Reliable and Interactive World Model was accepted to ICCV 2025.

I joined the Vision and Learning Lab at the University of Alberta as a postdoctoral fellow.

Selected Publications

full list

Motion-Guided Visual Tracking

Pengyu Zhang, Simiao Lai, Dong Wang, Huchuan Lu

Machine Intelligence Research, 2025

A motion-guided formulation for robust visual object tracking.

Robust Online Tracking With Meta-Updater

Jie Zhao, Kenan Dai, Pengyu Zhang, Dong Wang, Huchuan Lu

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023

code

A meta-updating strategy that learns when an online tracker should update its model.