Pushmeet Kohli
homepage:http://research.microsoft.com/en-us/um/people/pkohli/
search externally:   Google Scholar,   Springer,   CiteSeer,   Microsoft Academic Search,   Scirus ,   DBlife

Description

My research concerns the development of intelligent machines, with the particular aim of “teaching” computers to (1) understand the behaviour and intent of human users, and (2) to correctly interpret (or “See”) objects and scenes depicted in colour/depth images or videos. I work and publish in the fields of Computer Vision, Machine Learning, Discrete Optimization, Game Theory and Human-Computer Interaction. I am currently working on (1) the use of online social networks such as Facebook for conducting behavioural game theory experiments, and (2) the development of robust systems for human pose estimation and gesture recognition, and scene understanding using new depth sensors such as XBox KINECT.


Lectures:

tutorial
flag MAP inference in Discrete Models
as author at  British Machine Vision Conference (BMVC), Surrey 2012,
1196 views
  invited talk
flag Exploiting Problem Structure for Efficient Discrete Optimization
as author at  Discrete Optimization in Machine Learning,
549 views