Josephine Sullivan

	Josephine Sullivan Phone (office):+46 8 790 6136 Email : sullivan at nada.kth.se Address : KTH, CSC/CVAP, S-100 44 Stockholm, Sweden. I am a lecturer in computer vision within the School of Computer Science and Communication (CSC) at KTH, the Royal Institute of Technology, Stockholm, Sweden. I belong to the CVAP group and on a even more fine-grained classification I'm part of the Computer Vision Group within CVAP. My research efforts are devoted to the field of computer vision. In the past I have worked on the problems of visual tracking, human/object pose recognition and multi-target tracking. However, for the last couple of years I have been focusing on deep learning for solving computer vision problems.

Brief Bio	I obtained a D.Phil from Oxford University, Dept. Of Engineering Sciences in 2001 under the supervision of Prof. Andrew Blake. Prior to that I did a BA in Mathematics at Trinity College Dublin, Ireland. After my D.Phil studies I made my way to Sweden to work with Stefan Carlsson at KTH for a post-doc and now many years later I'm still here.

My Teaching & responsibilities at KTH	I'm director of the Master's programme in Machine Learning at KTH. Undergraduate Courses DD2427: Image Based Recognition and Classification (6 credits) Given every year in Period 4 DA2205: Introduction to the Philosophy of Science and Research Methodology (7.5 credits) Given every year in Period 1 and 2 in conjunction with the Philosophy Dept. Graduate Courses DD3364: Elements of Statistical Learning (9 credits) I gave this course in 2012/13. I may give it again if demand is sufficient. DD3001: Research: Theory, Method, Practice (7.5 credits) Given every year in Period 1 and 2.

Supervision	I have had the great pleasure to supervise and co-supervise several PhD students during my time at KTH. At the moment I co-supervise Hossein Azizpour Ali Sharif Razavian who are both working on deep learning. While the students below have successfully graduated Vahid Kazemi, (main supervisor) Thesis title: "Correspondence Estimation in Human Face and Posture Images", Graduated in 2014. Omid Aghazadeh, (secondary supervisor) Thesis title: "Data Driven Visual Recognition", Graduated in 2014. Heydar Maboudi, (secondary supervisor) Thesis title: "Improving Image Classification Performance using Joint Feature Selection", Graduated in 2014. Magnus Burenius, (main supervisor) Thesis title: "Human 3D Pose Estimation in the Wild using Geometrical Models and Pictorial Structures", Graduated in 2013. These students worked on a range of problems such as 3D human pose estimation and object detection

Presentations	Here are some of the talks I've given: Deep Convolutional Networks and solving large scale computer vision problems Presented at the Data Science Workshop at Stockholm University, Dec. 2014. (PDF) Deep Convolutional Neural Networks & Computer Vision Presented at the Machine Learning Meetup Stockholm about Deep Learning, June 2014. (PDF) Markerless Human Motion Capture from the Studio to Outdoor Environments Invited talk at Chalmers, September 2013. (PDF) Tracking and Labelling of Interacting Multiple Targets (PDF 17MB) Football Movies Common Football actions

Publications	Here are most of my publications in chronological order: Visual Instance Retrieval with Deep Convolutional Networks A.S. Razavian, J. Sullivan, A. Maki and S. Carlsson, arXiv preprint arXiv:1412.6574 [cs.CV], 2014. (PDF) Persistent Evidence of Local Image Properties in Generic ConvNets A.S. Razavian, H. Azizpour, A. Maki, J. Sullivan, C.H. Ek and S. Carlsson, arXiv preprint arXiv:1411.6509 [cs.CV], 2014. (PDF) From Generic to Specific Deep Representations for Visual Recognition H. Azizpour, A.S. Razavian, J. Sullivan, A. Maki and S. Carlsson, arXiv preprint arXiv:1406.5774 [cs.CV], 2014. (PDF) One Millisecond Face Alignment with an Ensemble of Regression Trees V. Kazemi and J. Sullivan, In Proc. IEEE Computer Vision and Pattern Recognition (CVPR 2014), Ohio, USA, June 2014. (PDF) CNN Features off-the-shelf: an Astounding Baseline for Recognition A. S. Razavian, H. Azizpour, J. Sullivan and S. Carlsson, CVPR workshop of DeepVision, June 2014. (PDF) Multi-view Body Part Recognition with Random Forests V. Kazemi, M. Burenius, H. Azizpour and J. Sullivan, In Proc. IEEE British Machine Vision Conference (BMVC 2013), Bristol, UK, Sept. 2013. (PDF) Discriminative tree-basd feature mapping M. Kobetski and J. Sullivan, In Proc. IEEE British Machine Vision Conference (BMVC 2013), Bristol, UK, Sept. 2013. (PDF) 3D Pictorial Structures for Multiple View Articulated Pose Estimation M. Burenius, J. Sullivan and S. Carlsson, In Proc. IEEE Computer Vision and Pattern Recognition (CVPR 2013), Portland, USA, June 2013. (PDF) Apprenticeship learning: transfer of knowledge via dataset augmentation M. Kobetski and J. Sullivan, In Proc. 18th Scandinavian Conf. on Image Analysis (SCIA 2013), Espoo, Finland, June 2013. (PDF) Improving Boosting Performance by Exclusion of Ambiguous Positive Examples M. Kobetski and J. Sullivan, In Proc. 2nd International Conf. on Pattern Recognition Applications and Methods (ICPRAM 2013), Barcelona, Spain, Feb 2013. (PDF) Adaptive Features for Object Classification H. Maboudi Afkham, S. Carlsson and J. Sullivan, In Series on Advances in Intelligent and Soft Computing, Springer, Volume 204, pages 121-133, 2013. Mixture Component Identification and Learning for Visual Recognition O. Aghazadeh, H. Azizpour, J. Sullivan and S. Carlsson, In Proc. 12th European Conf. on Computer Vision (ECCV 2012), 2012. (PDF) Using Richer Models for Articulated Pose Estimation of Footballers V. Kazemi and J. Sullivan, In Proc. IEEE British Machine Vision Conference (BMVC 2012), Guildford, England, Sept 2012. (PDF) Multi View Registration for Novelty/Background Separation O. Aghazadeh, J. Sullivan and S. Carlsson, In Proc. IEEE Computer Vision and Pattern Recognition (CVPR 2012), Rhode Island, USA, June 2012. (PDF) Face Alignment with Part-Based Modeling V. Kazemi and J. Sullivan, In Proc. IEEE British Machine Vision Conference (BMVC 2011), Dundee, Scotland, Sept 2011. (PDF) Novelty Detection from an Ego-Centric Perspective O. Aghazadeh, J. Sullivan and S. Carlsson, In Proc. IEEE Computer Vision and Pattern Recognition (CVPR 2011), Colorado Springs, June 2011. (PDF 12MB) Human 3D Motion Computation from a Varying Number of Cameras M. Burenius, J. Sullivan, S. Carlsson and K. Halvorsen, In Proc. IEEE Scandinavian Conference on Image Analysis (SCIA 2011), Ystad, Sweden, May 2011. (PDF) Automatic Learning and Extraction of Multi-Local Features O. Danielsson, S. Carlsson and J. Sullivan, In Proc. IEEE International Conference on Computer Vision (ICCV 2009), Kyoto, Japan, 2009 (PDF 9.7MB) Shading Models for Illumination and Reflectance Invariant Shape Detectors P. Nillius, J. Sullivan and A. Antonis, In Proc. IEEE Computer Vision and Pattern Recognition (CVPR 2008), Anchorage, June 2008 (PDF 2.2MB) Multi-Target Tracking -- Linking Identities using Bayesian Network Inference P. Nillius, J. Sullivan and S. Carlsson, In Proc. IEEE Computer Vision and Pattern Recognition (CVPR 2006), New York City, June 2006 (PDF 0.5MB) Tracking and Labelling of Interacting Multiple Targets J. Sullivan and S. Carlsson, In Proc. 9th European Conf. on Computer Vision (ECCV 2006), 2006. (PDF 4.8MB) Monocular 3D Reconstruction of Human Motion in Long Action Sequences G. Loy, M. Eriksson, J. Sullivan and S. Carlsson, In Proc of European Conference on Computer Vision (ECCV 2004), 2004. (PDF 1.9MB) Pose-Based Clustering in Action Sequences G. Loy, J. Sullivan and S. Carlsson, Workshop on Higher-Level Knowledge in 3D Modeling & Motion Analysis, 2003. (PDF 1.2MB) Recognizing and Tracking Human Action J. Sullivan and S. Carlsson. In Proc of European Conference on Computer Vision (ECCV 2002), 2002. (PDF 591KB) Bayesian Correlation J. Sullivan, A. Blake, M. Isard and J. MacCormick, In International Journal of Computer Vision, 44(2), pages 111-135, 2001 Action Recognition by Shape Matching to Key Frames S. Carlsson and J. Sullivan, Workshop on Models versus Exemplars in Computer Vision, 2001. (PDF 180KB) Guiding Random Samples by Deterministic Search J. Sullivan and J. Rittscher, In Proc. International Conference on Computer Vision (ICCV 2001), Vancouver, Canada, 2001. (gzipped PS 2MB) Statistical Foreground Modelling for Object Localisation J. Sullivan, A. Blake, and J. Rittscher. In Proc. European Conference on Computer Vision (ECCV 2000), Dublin, Ireland, May 2000. (gzipped PS 2.2MB) Object Localisation by Bayesian Correlation J. Sullivan, A. Blake, M. Isard, and J. MacCormick. In Proc. International Conference on Computer Vision (ICCV 1999), Corfu, Greece, Sept. 1999. (gzipped PS 2.2MB)

josephinesullivan