This article is an illustrated summary of a recent paper we presented at CVPR 2009.  We leverage some of the linear properties of optical flow fields to develop a method that automatically learns the relationship between camera motion and optical flow from data.  The method can handle arbitrary imaging systems including very severe distortion, curved mirrors, and multiple cameras.  Using this method, a robot can estimate it's motion in real time from video while detecting "motion anomalies" such as nearby or moving objects.

There was a very interesting plenary talk at ICRA 2009 about "Computational Cameras" given by Prof. Shree Nayar of Columbia University.  A video of the plenary is included below, as well as a discussion of some of its contents -- from assorted pixel techniques for high dynamic range to flexible depth of field photography -- all very cool stuff!  These developments are particularly relevant to robotics, as cameras are probably the most ubiquitous sensors encountered.  This video was made available in the ICRA 2009 podcasts.  While there is a large push for open-access journals / conferences, freely-available recordings of conference talks is even more lacking.  As I find these more entertaining than television, I really hope this becomes a common trend (perhaps the RSS committee members are watching...?).