Deploying Computer Vision to Help Social Distancing Amid Pandemic Outbreak

As the pandemic crisis worsens on the horizon and the civilization fears to be on brink of extinction, human developed tools are proving to be our best asset. Computer vision is one such brainchild. Zensors, a Carnegie Mellon University startup is one of the firms that is stepping up the game. It enables smart and reactive spaces with better crowd screening through cutting-edge computer vision technologies. This Computer vision activity monitoring startup developed a suite of tools that uses machine learning. It is specifically designed for businesses like airports, offices and retail environments to track things like identifying and dynamically dispose of pick-up trash, restaurant occupancy, estimate lines. And now to put the icing on the cake, over the past few weeks, their team across Pittsburgh, New York City, and Mumbai are trying to determine how the platform can be helpful in response to COVID-19.

Founded in 2018, Zensors was among the earliest adopters of computer vision in the field of valuable data extraction. Now with the pandemic affecting people from all walks of life, this can help to monitor physical and social distancing among the public. The software can count the number of people in a line or an area, then calculate distances between people in the said space. Based on this data, it will send an email or SMS alert in case protocols are being violated. This may seem an easy computation. But years back using security feeds to come up with or accomplish that, was not possible.

This can help to:

· Know the number of people in given public place or facility

· If the gatherings are confined by mandated congregation limit

· Know where and when the cleaning personnel should focus their activities of sanitizing and waste disposal

· Check if people are wearing face masks in the suggested regions

· Observe if people are following recommended social distancing policies.

There won't be any necessity to install extra CCTVs as Zensors utilizes the available ones as IoT sensors to provide anonymous real-time occupancy data. The platform supports ONVIF, HTTPS, FTP, RTSP and several proprietary camera protocols. Given its reputation, many airports had approached it regarding applying the technology to public health considerations. Rather than profiting off them, they decide to provide the platform for free until 1 June. It is also inviting machine learning researchers and experts to collaborate on using the platform.

During this unprecedented crisis, the team also expressed their interest in working with geographic information system (GIS) experts who can help to integrate collected data onto ESRI and other mapping platforms. They provide no-cost, expedited setup and one free Zensors On-Premise Interface (ZOPI) hardware device to ensure secure, encrypted connectivity to the Zensors cloud-based servers hosted by Amazon Web Services. Furthermore it uses stringent protocols to ensure the safety and integrity. If desired facial blurring can be enabled to prevent capture of identifiable information.