Welcome to the homepage for KTH's Computer Vision Group! We are part of the CVAP laboratory in the School of Computer Science and Communication at the Royal Institute of Technology, Stockhom, Sweden.
At the Computer Vision Group we perform research in the field of computer vision and machine learning. We have a special emphasis on scalable methods as we want to do vision on data from wearable cameras - recognising what we've seen, have we seen it before, is it different from what I've seen before. Currently, our main focus is exploiting machine learning and models to perfom object recognition and detection and human pose estimation.
To discover what we are currently thinking about in our research check out our Reading Group page.
- Heydar successfully defended his thesis in May 2014. We wish him all the best with his new job as a post-doc in Computational Biology at KTH.
- We have created a webpage to keep track of the latest results for using Deep Learning to learn visual features:
Benchmark of Deep Learning Representations for Visual Recognition
- Ali and Hossein's paper CNN Features off-the-shelf: an Astounding Baseline for Recognition has been accepted to the Deep Learning Workshop at CVPR 14.
- Ali's paper Estimating Attention in Exhibitions Using Wearable Cameras has been accepted to ICPR 14.
- Vahid's paper One Millisecond Face Alignment with an Ensemble of Regression Trees has been accepted to CVPR 14.
- Magnus successfully defended his thesis in January 2014. We wish him all the best with his new job at the start-up company Volumental.
- We are very happy that our paper Multi-view Body Part Recognition with Random Forests, was awarded The Best Industry Paper Prize at BMVC 2013.