Joseph Shtok           

In the field of Computer Vision, my two main domains of expertise are the object recognition algorithms based on Deep Learning, and the algorithms for 3D pose detection and tracking of objects and scenes. The former represent a doorway to the realm of Artificial Intelligence, while the latter serve as the algorithmic core for the Augmented Reality applications. Recently, I am focusing on Few-Shot learning, that is, learning from few labeled examples per category. 

Information about our team's activity, including the recent papers and the IMVC 2019 presentations, can be fount in the CVAR's home page http://www.research.ibm.com/haifa/dept/imt/ist_dm.shtml