The access control provided by a physical lock is based on the assumption that the information content of the corresponding key is private --- that duplication should require either possession of the key or a priori knowledge of how it was cut. However, the ever-increasing capabilities and prevalence of digital imaging technologies present a fundamental challenge to this privacy assumption. Using modest imaging equipment and standard computer vision algorithms, we demonstrate the effectiveness of physical key teleduplication --- extracting a key's complete and precise bitting code at a distance via optical decoding and then cutting precise duplicates. We describe our prototype system, Sneakey, and evaluate its effectiveness, in both laboratory and real-world settings, using the most popular residential key types in the U.S.

We present a scalable approach to recognizing and describing complex activities in video sequences. We are interested in long-term, sequential activities that may have several parallel streams of action. Our approach integrates temporal, contextual and ordering constraints with output from low-level visual detectors to recognize complex, long-term activities. We argue that a hierarchical, object-oriented design lends our solution to be scalable in that higher-level reasoning components are independent from the particular low-level detector implementation and that recognition of additional activities and actions can easily be added. Three major components to realize this design are: a dynamic Bayesian network structure for representing activities comprised of partially ordered sub-actions, an object-oriented action hierarchy for building arbitrarily complex action detectors and an approximate Viterbi-like algorithm for inferring the most likely observed sequence of actions. Additionally, this study proposes the Erlang distribution as a comprehensive model of idle time between actions and frequency of observing new actions. We show results for our approach on real video sequences containing complex activities.

This paper provides a practical framework for planning and control of formations of multiple unmanned ground vehicles to traverse between goal points in a dynamic environment. This framework allows online planning of the formation paths using a Dijkstra's search algorithm based on the current sensor data. The formation is allowed to dynamically change in order to avoid obstacles in the environment while minimizing a cost function aimed at obtaining collision-free and deadlock-free paths. Based on the feasible path for a virtual leader of the group, the trajectory planner satisfies the kinematic constraints of the individual vehicles while accounting for inter-vehicle collision and path constraints. A Lyapunov based controller is designed to keep the vehicles on their planned trajectories. Illustrative simulations of groups of unmanned ground vehicles and their laboratory implementation with three unmanned ground vehicles are presented.