- Voxel51 has developed a Physical Distancing Index tool that assigns a score which helps people understand and compare how the coronavirus is changing social behaviors
Video data scientists at Ann Arbor, Michigan-based Voxel51 have launched an interactive AI-powered tool that enables users to watch live street cameras and explore how COVID-19 affects social behavior in real-time. The Voxel51 Physical Distancing Index (https://pdi.voxel51.com/) assigns a score that helps people understand and compare how the coronavirus is changing social behaviors.
Voxel51’s video analysis technology is currently gathering and analyzing historical and real-time video streams from public street cameras in London, New York City, Fort Lauderdale, New Jersey, Prague and Dublin, Ireland. And the company’s computer vision and deep learning models are able to detect the number of pedestrians, vehicles, and other human-centric objects in each live stream. The PDI score (or the average amount of human activity over the previous 24 hours) is automatically calculated every 15 minutes for tracking the change over time.
The PDI scores are represented in every time-stamped data point on the chart along with relevant COVID-19 news headlines from the day and a snapshot image from the live video scene.
And users can also compare scores across locations to see the differences in PDI relative to each city over time. The PDI dashboard is updated in real-time and will continue to integrate live street cams from additional cities around the world.
“The coronavirus has had an intense impact on daily life across the globe, and we wanted to find a way to use our technology as a tool for public awareness,” said Voxel51 co-founder and CEO Jason Corso. “We were able to apply our computer vision methods to compare how people are responding to physical distancing measures around the world and to study how exposure rates correlate with news on the coronavirus and physical distancing mandates over
The PDI tool is non-invasive and does not extract any identifying information about the individuals in the video or use other data sources like mobile phone signals — which allow only approximate location estimates. Voxel51 focuses on specific centralized locations of interest in each city and watches the changes in public social activity in real-time.
“The data in the graphs speaks for itself. Coronavirus and the necessary preventative measures across the globe have had an intense impact on daily life,” added Voxel51 co-founder and CTO Brian Moore. “All of the cities we’re tracking have seen a sharp decline in public activity during the month of March.”
