My name is Carl Henrik Ek and I am a post-doctoral researcher in Computer Vision and Machine Learning at the International Computer Science Institute (ICSI) in Berkeley California. I am also affiliated with the Electrical Engineering and Computer Science (EECS) department at the University of California, Berkeley as a visiting post-doc. I am a member of Professor Trevor Darrells research group.

Before moving to California I did my PhD in the United Kingdom at the University of Oxford Brookes. My supervisors where Dr. Neil Lawrence at the University of Manchester and Professor Phil Torr at Oxford Brookes. Prior to this I got a MSc in Vehicle Engineering from KTH in my native Sweden during which I spent a year at the University of Bristol.

Publications

[1] C. H. Ek, P. H. Torr, and N. D. Lawrence. Gaussian process latent variable models for human pose estimation. In 4th Joint Workshop on Multimodal Interaction and Related Machine Learning Algorithms (MLMI 2007), volume LNCS 4892, pages 132–143, Brno, Czech Republic, Jun. 2007. Springer-Verlag. [PDF]

[2] C. H. Ek, P. H. Torr, and N. D. Lawrence. Ambiguity modeling in latent spaces. In Proceedings of the International Workshop on Machine Learning for Multimodal Interaction (MLMI’08), number 5237 in Lecture Notes in Computer Science, pages 62–73, 2008. [PDF]

[3] C. H. Ek, P. H. Torr, and N. D. Lawrence. Gp-lvm for data consolidation. Learning from Multiple Sources, NIPS, 2008.[PDF]

[4] C. H. Ek, P. Jaeckel, N. Campbell, N. Lawrence, and C. Melhuish. Shared Gaussian Process Latent Variable Models for Handling Ambiguous Facial Expressions. In AIP Conference Proceedings, volume 1107, pages 147–153, 2009. [PDF]

Talks

Code

An important part to make progress in Computer Science is the publication of code to make it possible for others to recreate published results. The code for the publications above is publicly available through my supervisors webpage[URL].

Departmental Seminar - CVAP KTH Stockholm

Introduction to Shared Gaussian Process Latent Variable Models. [slides]

Department of Mathematics - University of Manchester

Introduction to the GP-LVM and some of its extensions given at the EMBody - Training Week at the Maths Department at University of Manchester. [slides]

Departmental Seminar - University of Bristol

A brief tutorial on Gaussian Processes followed by an introduction to the GP-LVM. We then present our work on using Shared GP-LVM model. The models are demonstrated by modeling a human pose estimation from silhouette task. [slides]

Department of Engineering Science - University of Oxford

Tutorial on spectral methods for dimensionality reduction. It goes through quite a few algorithms and talks about the disadvantages and advantages. I also tried to explain them in the same framework thereby making the connection and similarities between them clearer.  [slides] [code]

Department of Computer Science - University of Manchester

Brief tutorial on Markov Random Fields and how they can be cast as Flow Network making efficient inference possible [slides]