Multi-camera vehicle counting using edge-AI

This paper presents a novel solution to automatically count vehicles in parking lot using images captured by smart cameras. Unlike most of the literature on this task, which focuses analysis single images, proposes use multiple visual sources monitor wider area from different perspectives. The proposed multi-camera system is capable estimating number cars present entire directly board edge devices. It comprises an on-device deep learning-based detector that locates and counts decentralized geometric-based approach can analyze inter-camera shared areas merge data acquired all We conducted experimental evaluation extended version CNRPark-EXT dataset, collection taken campus National Research Council (CNR) Pisa, Italy. show our robust takes advantage redundant information deriving cameras, improving overall performance without requiring any extra geometrical monitored scene.