Sam Johnson

Sam Johnson

PhD Student

Computer Vision Group

School of Computing, University of Leeds

United Kingdom, LS2 9JT

E-mail Updated: sam.johnson@crl.toshiba.co.uk


About Me

After graduating from the University of Leeds with a first class BSc (hons) in Computer Science I chose to undertake a Ph.D. in the Computer Vision Group supervised by Dr. Mark Everingham.


Research

Through my research I am aiming to develop computer vision methods for estimating the pose of humans present in images taken in uncontrolled environments, with unknown lighting, anatomy, clothing, etc. Success would contribute to all image understanding applications involving humans, such as searching photographs, human-robot interaction, interpreting CCTV footage, automatically narrating video for visually impaired viewers, etc.


Publications

Automatically Improved Annotation S. Johnson and M. Everingham
"Learning Effective Human Pose Estimation from Inaccurate Annotation"
In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR2011)
PDF Dataset (zip)
Estimated Pose S. Johnson and M. Everingham
"Clustered Pose and Nonlinear Appearance Models for Human Pose Estimation"
In Proceedings of the 21st British Machine Vision Conference (BMVC2010)
PDF Dataset (zip)
Estimated Pose S. Johnson and M. Everingham
"Combining Discriminative Appearance and Segmentation Cues for Articulated Human Pose Estimation"
In Proceedings of the 2nd IEEE International Workshop on Machine Learning for Vision-based Motion Analysis (MLVMA09), in conjunction with ICCV2009.
PDF Poster (Presented at BMVA Summer School 2009, Winner of best poster award)

Theses

Articulated Human Pose Estimation in Natural Images S. Johnson
"Articulated Human Pose Estimation in Natural Images"
Doctoral thesis, University of Leeds, October 2012.
PDF
Video Super-resolution from Scene Specific Images S. Johnson
"Video Super-resolution from Scene Specific Images"
Undergraduate thesis, University of Leeds, June 2008.
Winner of the Buckley Prize for best undergraduate research project.
PDF