Hello There!



PhD Student at CVAP, CAS, CSC - KTH Royal Institute of Technology
Room 609, Teknikringen 14, 114 28 Stockholm, Sweden
Phone: +46 8 790 7224
Email: martinhjelm -- k t h . se
Blog: The NonConditional Beast
Github: github.com/MartinHjelm
Google Scholar: scholar.google.com
Curriculum vitae(CV): CV(pdf)

About Me

Im a PhD student at CVAP, CSC - KTH Royal Institute of Technology. My main research interest are machine learning, computer vision and robotics.

I have a blog on machine learning related topics: The NonConditional Beast and a blog on programming related issues at martinhjelm.github.io/.

Other interests are competing at Kaggle, learning to play as many instruments as possible, anthropology, architecture, food, long distance running, the Arts. The list could go on for a while. I also do a lot of web-programming like most programming literate person I like startups. I created and designed Bookle, a site for selling second hand course literature and I have done several other web projects as a consultant and by myself.


(Videos best viewed at 1.5 speed!)

Learning Human Priors for Task-Constrained Grasping

M. Hjelm, ,C. H. Ek, R. Detry, and D. Kragic, In 10th International Conference Computer Vision Systems, ICVS, 2015.

Representations for Cross-task, Cross-object Grasp Transfer

M. Hjelm, R. Detry, C. H. Ek and D. Kragic, In IEEE International Conference on Robotics and Automation, 2014.

Sparse Summarization of Robotic Grasping Data

M. Hjelm, C. H. Ek, R. Detry, H. Kjellstrom and D. Kragic, In IEEE International Conference on Robotics and Automation, 2013.



TA - DD2431 Machine Learning
TA - DD2423 Image Analysis and Computer Vision


TA - DD1346 Object-Oriented Program Construction,
TA - DD2427 Image Based Recognition and Classification

Past experiences

I did my Master Thesis on image segmentation at TTI(Toyota Technological Institute) in Chicago under supervision of Raquel Urtasun. Prior to that I was at TU Berlin studying machine learning courses under professor Mannfred Opper and professor Klaus-Robert Muller. I did my undergrad studies and part of my master studies at KTH, at the Engineering Physics section focusing on numerical methods for differential equations.


You can find all my code for various project at GitHub.