Physical Distancing and Crowd Density Monitoring
Fig . 5 : This first set of heatmaps summarizes the locations of every detected person aggregated over 5-minute intervals , answering questions about the areas of highest traffic at different times . On the left , we aggregate over 11:30am to 11:35am whereas on the right , we aggregate over 12:30pm to 12:35pm . We observe different spatial patterns of activity between the twotime intervals , such as the increased traffic around the cashiers near the bottom of the image . We also observe similarities such as the high traffic around the horse-shoe shaped salad bar area or the primary cashier station at the top-left . In practice , we build heatmaps over all timeintervals of interest and animate to show their continuity .
Sometimes , despite our best efforts , there can be areas in an environment where it is nearly impossible to distance , such as a narrow hallway , or a blocked sidewalk , and so on . A welldesigned space can allow high traffic while maintaining physical distancing . But at the other end of the spectrum , a poor-design can thwart our best efforts to physical-distance . The first heat map only studies the overall traffic volume and cannot answer where physical distancing is failing .
IIC Journal of Innovation - 11 -