ICCV 2013 Tutorial on Part-based Models for Recognition

Time: December 1, 2013 9am to 1pm
Location: ICCV 2013, Sydney, Australia


Tutorial description

The tutorial will provide a brief overview of the history of part-based models and their applications to a variety of recognition problems such as detection, pose estimation, attribute and action recognition. These include the early models that are variants of pictorial structures, Hough voting based methods such as implicit shape models, and more recent methods based on learning parts using tree structure latent variable models (DPM), supervised methods such as Poselets, etc. We will also highlight some challenges and future directions in this area.