Know exactly when employees perform tasks and how long they take. Was nightly maintenance done on that critical piece of machinery? At what time? Did it take the usual 20 - 25 minutes or did it take 2 minutes? Turning video into answers to these questions can prevent problems due to neglected tasks before they arise, and identify opportunities to better train employees.
Inventory shrinkage is a multi-billion dollar problem in the retail industry. Most video cameras in retail spaces are empty threats of the possibility of being caught, with the relevant video being pulled only after a significant loss is discovered if it hasn’t already been overwritten. We provide tools that automatically detect suspicious activity in streaming video and send either real time alerts or daily/weekly digests, complete with the relevant snippets of video.
Lots of video contains long periods of nothing much happening, or the same thing happening over and over, followed by a transition to something else. Imagine being able to turn streaming video from a home webcam into a quick 3 - 4 minute video diary of what your child did today with the babysitter/nanny or at daycare.
In this video we show a number of examples of the kinds of events that you can teach a computer to find automatically with Scrutable technology. Each of the examples is the end product of using Scrutable tools to teach the computer to find specific kinds of activities in real-world videos. Click here for a Demo
People are really good at understanding what’s going on in audio and video recordings. We hear a crashing sound in a restaurant recording and know that someone dropped a dish. We see a surveillance video in which someone slips a watch into their pocket and know they intend to steal it. But there are not enough eyes and ears for all of the video currently being produced. At Scrutable AV we’re bridging that gap by applying deep knowledge of both the retail industry and machine learning to allow machines to understand audio and video in ways that can make a real impact on the bottom line. We do that by develop event detection solutions that are easy to train, easy to maintain, fast, and accurate.
Dr. Tim Oates is an Oros Family Professor of Computer Science in the CSEE Department at the University of Maryland, Baltimore County, and a co-founder of Scrutable AV. He has a Ph.D. in Computer Science from the University of Massachusetts, Amherst, and spent a year as a postdoc in the AI Lab at MIT. He has more than 25 years of research experience in AI, Machine Learning, and Data Mining, and extensive data science consulting experience in industries as diverse as construction, publishing, defense, and health care.
Beenish Bhatia is the Co-founder of Scrutable AV and holds a Masters degree in Computer science from the University of Maryland, Baltimore County. Entrepreneurship is nothing new to him and is the owner of multiple stores with multiple franchise systems and a trucking company in the USA. He has immense expertise in the field of retail and logistics industry. His knowledge in the fields of AI, Machine learning and Data mining has led to the logical birth of Scrutable AV.
Learn more about our technology and current use cases, or explore how our audio and video understanding technology can help your bottom line.