In this paper, we propose to use commercial off-the-shelf (COTS) monostatic\nRFID devices (i.e. which use a single antenna at a time for both transmitting\nand receiving RFID signals to and from the tags) to monitor browsing activity\nof customers in front of display items in places such as retail stores. To this\nend, we propose TagSee, a multi-person imaging system based on monostatic RFID\nimaging. TagSee is based on the insight that when customers are browsing the\nitems on a shelf, they stand between the tags deployed along the boundaries of\nthe shelf and the reader, which changes the multi-paths that the RFID signals\ntravel along, and both the RSS and phase values of the RFID signals that the\nreader receives change. Based on these variations observed by the reader,\nTagSee constructs a coarse grained image of the customers. Afterwards, TagSee\nidentifies the items that are being browsed by the customers by analyzing the\nconstructed images. The key novelty of this paper is on achieving browsing\nbehavior monitoring of multiple customers in front of display items by\nconstructing coarse grained images via robust, analytical model-driven deep\nlearning based, RFID imaging. To achieve this, we first mathematically\nformulate the problem of imaging humans using monostatic RFID devices and\nderive an approximate analytical imaging model that correlates the variations\ncaused by human obstructions in the RFID signals. Based on this model, we then\ndevelop a deep learning framework to robustly image customers with high\naccuracy. We implement TagSee scheme using a Impinj Speedway R420 reader and\nSMARTRAC DogBone RFID tags. TagSee can achieve a TPR of more than ~90% and a\nFPR of less than ~10% in multi-person scenarios using training data from just\n3-4 users.\n