Advancements in Monocular Human Pose Estimation and Motion Recognition

Project: Internal Research Project

Project Details

Fund Amount (RMB)



This research project aims to create an advanced computer program to understand how people move by looking at a regular picture or video. Imagine taking a single photo of someone and instantly knowing not just where their body parts are in three dimensions but also what kind of movement they're doing, like running or waving. We want to achieve this using the power of deep learning, a type of artificial intelligence. Our method uses a special kind of neural network inspired by the human brain to learn from lots of pictures and become good at recognising people's poses and actions. We've already made significant progress, outperforming previous methods when tested on an extensive dataset. This technology has the potential to be used in various fields like healthcare, sports analysis, and even in making video games more realistic. Our research will explore the possibilities and challenges ahead, laying the groundwork for exciting future discoveries in this area.
Project CategorySchool Research Priorities Seed Funding (SRPSF)
Effective start/end date1/12/2329/11/24


  • Human Motion Analysis, Pose Recognition, Deep Learning, Computer Vision


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.