Computer vision technologies in the post-COVID era

With machine learning & artificial intelligence makes progress with leaps & bounds, we humans aim to leverage it to the fullest to solve complex problems with ease & make our lives easy.

Data, not just in the form of numbers but also digital images & videos are being leveraged to the fullest. Extracting the features of an image or a video & analyzing it can lead to meaning full insights that are increasingly finding its usage in various cases.

Computer vision technology, as they say, it today is now core to many disruptions like self-driving cars, robotics applications, etc. I was fortunate to work on finding business-relevant use cases for such a solution. We explore different areas where such a technology can be employed like identifying customer experience using the visual cues from their facial gestures, claims assessment for insurance, quality control for line robots, predictive asset maintenance, etc.

But in these changing times, one very interesting use case of computer vision technologies that are catching eyeballs of the corporates & gaining acceptance across sectors is its capability to analyze crowd density in a particular area. Such technologies have now become an area of prime interest esp. to the government bodies to implement crowd management strategies across various public places. Safety in working areas & maintaining the crowd flow be it in busy production lines or retail stores also builds a strong case for the usage of computer vision technologies.

There are a variety of features that businesses can include ranges from headcounts, crowd density, facemask detection, path detection, intrusion activities, customer in-store time analysis & queue management. The solutions providers should have the capability to tailor their offerings according to the clientele they are pitching it to.

However, there are some infrastructural roadblocks that the solution providers should focus on to come up with the right solution. These type of solutions needs to be applicable even in a low internet connectivity area with a near real-time feed that is duly encrypted & follows the security norms. Overcome these challenges & you have a near-perfect solution for crowd monitoring & management.

Pre-COVID estimates suggested that the market for computer vision technologies is expected to reach about 19Bn Euros by 2025. But the figures, are going to change now, up or down, totally depends on the successful delivery of the promises that this technology has made.