A novel software has been developed by the researchers of IISc (Indian Institute of Science) which will help through apps and algorithms to track and analyze video feeds from cameras placed across cities.

IISc released a statement, "Such analysis is not only useful for tracking missing persons or objects, but also for 'smart city' initiatives such as automated traffic control", as many cities across the world have set numerous video cameras. The institute further added that the machine learning models can rummage for a specific purpose only through the feed from such cameras e.g. to track a stolen car. These models need a software platform, similar to a computer operating system, to run as they cannot work by themselves.

Associate Professor of IISc Department of CDS (Computational and Data Sciences) Yogesh Simmhan said, "There's been a lot of research on increasing the accuracy of these models, but sufficient attention hasn't been paid to how you make the model work as part of a large operation". To address this problem, a software platform 'Anveshak' has been developed by Simmhan's lab. The software can not only run tracking models proficiently but can also plug in advanced vision tools of computer which can adjust to different parameters intelligently and in real-time.

IISc further informed, "A key feature of the platform is that it allows a tracking model or algorithm to focus only on feeds from certain cameras along an expected route and tune out other feeds. It can also increase or decrease, automatically, the search radius or 'spotlight' based on the object's last known position".

In 2019, during the IEEE TCSC SCALE Challenge Award, Simmhan's lab showed, as a part of winning entry, how Anveshak could be utilized to control the traffic signals and also to automatically show 'green routes' for ambulances, in case of emergency. It also engaged a 'spotlight tracking algorithm' to control which feeds needed to be analyzed.

Simmhan's lab is now dealing with integrating privacy restrictions within the software and they are also working on techniques to use the software to track multiple objects at the same duration.