University of Cambridge > Talks.cam > CUED Computer Vision Research Seminars > Long-term tracking of human pose in videos

Long-term tracking of human pose in videos

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact gd348.

This talk presents random forest and convolutional neural network (ConvNet) based methods for real-time human pose estimation in videos. I will show the proposed methods accurately track the 2D position of upper body joints, such as the wrists and elbows, despite fast motion and continuously changing cluttered background. The main focus of this talk is the introduction of my recently developed personalised ConvNet pose estimation method, which greatly outperforms the current state-of-the-art on numerous video datasets. Furthermore, I will introduce a new challenging video pose dataset collected from YouTube, and show how we can generate copious amounts of labelled pose training data efficiently and fully automatically.

This talk is part of the CUED Computer Vision Research Seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2024 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity