In this paper we present SCOPES, a distributed Smart Camera Object Position Estimation sensor network System that provides maps of distribution of people in indoors environments. Each node in the system is comprised of a cyclops camera that performs local detection and processing of the visual information and a tmote sensor node, which provides multihop communication. SCOPES uses local adaptive techniques that enables intelligent duty-cycling 1 between the active sensing and the information processing tasks performed at each node. The system switches between the fast and simple background subtraction algorithms for object detection, to the more computationally intensive object grouping algorithms for estimating the number and direction of travel of multiple persons in the local field of view. By aggregating meta-information generated by each node, SCOPES is able to minimize the total data transmitted in the network and still be able to generate an accurate density estimation map of the coverage area.Using analysis, simulation and experimentation, we show that the system is able to provide a small global error estimate of the spatio-temporal distribution of people in indoors environments despite the absence of continuous sensing when doing local information processing and sparse coverage. Moreover, we show that performance of the system degrades gracefully in the presence of memory and people mobility constraints. In the paper we show the results of people density estimation, power consumption, memory usage, latency and detection probability on a real system deployment comprised of 16 nodes running in a research building at the University of California, Merced.