Computer vision is an invaluable tool for analyzing large amounts of video footage, especially for applications like law enforcement. Assistant Professor Yogesh Rawat, a researcher at the UCF Center for Research in Computer Vision, is working on software that can address privacy concerns in video surveillance. Supported by funding from the U.S. National Science Foundation’s Accelerating Research Translation program, Rawat’s software can obscure identifiable information, such as faces and clothing, in real time by adding perturbations to the RGB pixels in the video feed.
The main goal of Rawat’s work is to develop technology that can be implemented on edge devices, such as drones and public surveillance cameras, to analyze video footage in real time. This involves creating algorithms that can process data quickly and efficiently, without requiring large amounts of computing power. This technology has the potential to benefit a wide range of users, from nursing homes and childcare centers to authorities using surveillance cameras.
With funding from the NSF ART program, Rawat aims to identify potential users for his technology and ensure that it can be implemented effectively in various settings. His work builds on previous projects undertaken by other CRCV members, including founder Mubarak Shah and researcher Chen Chen, which have focused on analyzing security videos, training AI models on a smaller scale, and developing software for the detection of multiple actions and objects of interest.
Rawat’s motivation for working in computer vision comes from a desire to improve the world we live in by leveraging visual perception technology to make tasks easier and more efficient for humans and society as a whole. By developing software that can process video footage quickly and accurately, Rawat hopes to facilitate better surveillance practices while maintaining privacy and security for individuals recorded on camera.
The technology being developed by Rawat has the potential to revolutionize video surveillance practices by enabling real-time analysis of footage and protecting the privacy of individuals captured on camera. By ensuring that the software is efficient, fast, and adaptable to various edge devices, Rawat is paving the way for a new era of surveillance technology that balances the need for security with respect for privacy rights. With ongoing funding and support, Rawat’s work has the potential to make a significant impact in various industries and sectors that rely on video surveillance for monitoring and security purposes.
