Approximate Inference using Monte Carlo Methods Frank Dellaert College of Computing, Georgia Institute of Technology Sampling can be used as an approximate inference method in graphical models, and is becoming increasingly popular in computer vision and robotics. Importance sampling in particular has seen enormous success as the main inference engine in the Condensation algorithm, and in particle filters in general. However, a more powerful methodology is Markov chain Monte Carlo, which was discovered by Metropolis some 50 years ago. I will motivate MCMC from a Markov chain perspective, and point out several applications in vision and robotics where MCMC can be used. As a bonus feature, after my tutorial, you will never look at Google in quite the same way. Frank Dellaert is an Assistant Professor at the College of Computing, Georgia Institute of Technology. He graduated in 2001 with a Ph.D. from Carnegie Mellon University. His research focuses on probabilistic methods in Robotics and Computer Vision: he has applied Markov chain Monte Carlo sampling methodologies in a variety of novel settings, most notably to address the correspondence problem in computer vision. Before that, with Dieter Fox and Sebastian Thrun, he has introduced the Monte Carlo localization method for estimating and tracking the pose of robots, which is now a standard and popular tool in mobile robotics. Since coming to Georgia Tech, he explored the theme of probabilistic, model- based reasoning paired with randomized approximation methods in three main research areas: Advanced sequential Monte Carlo methods, Spatio-Temporal Reconstruction from Images, and Simultaneous Localization and Mapping. His homepage can be found at http://www.cc.gatech.edu/~dellaert