High-quality continuous counts of pedestrian volume are necessary to evaluate the effects of pedestrian infrastructure investments and to improve pedestrian volume modeling. Automated pedestrian counting devices can meet the need for continuous counts of pedestrian volume and reduce the labor cost associated with manual pedestrian counting and data entry. However, most existing automated pedestrian devices are not well suited to the task of counting pedestrians in outdoor environments, and little is known about their effectiveness and accuracy. This study addresses the lack of performance information on automated counting devices by providing a review of commercially available devices and by testing the accuracy of a promising device in an outdoor urban context. It finds that a dual sensor passive infrared device is capable of producing reasonably accurate pedestrian volume counts in the outdoor urban context. It also finds a high degree of inter-reliability between counts collected by field observers and through video recordings.