VSBLTY On the Edge
The vast potential of the VSBLTY software suite hinges on the processing capabilities of Edge.
On the edge, VSBLTY can bring the complexity of machine learning down to the level of the computer, without a dependency on sending information out to the cloud via a network or internet connection.
The result is a unique and wildly powerful combination of real-time data collection and targeted interactive content.
The OpenVINO™ Toolkit
Intel’s OpenVINO platform for computer vision is fundamental to the performance of VSBLTY’s advanced machine learning. OpenVINO enables VSBLTY’s algorithms to run at lightning speed, with incredible precision, and at scale.
Through OpenVINO, VSBLTY can maintain its services on the edge, in the cloud, or a hybrid of both. With this versatility, machine learning models can be hot-swapped in or out anytime, without needing to reprogram.
Utilizing the advantages of edge computing, VSBLTY is able to seamlessly combine its software with proprietary optics to deliver groundbreaking digital display, real-time data at point of engagement, and uniquely effective security and surveillance applications.
Advanced Object Recognition
VSBLTY’s computer vision system recognizes common objects and reacts to what the camera “sees” in real-time. Digital displays in retail can change content based on what is happening in the surroundings, while in security, persons or objects of interest can be identified instantly, with the necessary authorities being alerted.
Analytics and Measurement of Retail
VSBLTY provides a new standard in analytics and measurement in retail by combining computer vision with interactive digital signage. By commanding guests’ attention with dynamic digital displays, custom content can be provided based on what the camera in the display “sees”, while gaining valuable insights in real-time.
Machine Learning
VSBLTY utilizes the power of computer vision through machine learning. Machine learning gives computers the ability to learn and react without being directly programmed. With computer vision, video images can be processed and analyzed in real-time, producing actionable information about those images.
Intersection of Marketing and Security
VSBLTY’s computer vision system can react to whatever is going on in the environment, in real-time, through common digital signage interfaces. Both people and objects can be identified, with the appropriate response being initiated automatically, whether for marketing or security purposes.
NIST Validation: VSBLTY Built Using Most Accurate Facial Recognition Algorithms
What is NIST?
NIST is the United States Commerce Departments’ National Institute of Standards and Technology. With cyber-attacks becoming more universal and complex, there was a need to create security standards for the way companies and organizations use techniques, languages and rules to combat security threats. To meet this challenge, NIST created a set of cybersecurity guidelines for private sector organizations to follow in order to be better prepared to identify, detect, and respond to cyber-attacks.
NIST Facial Recognition Vendor Test
NIST’s recent face recognition/person identification evaluation of 127 software algorithms from 39 developers found that Facial Recognition (FR) software got 20 times better at searching databases to find matching photographs between 2014 and 2018. NIST judged how well an algorithm could match a person’s photo with a different one, of the same person, stored in a large database. This type of “one of many” search is often employed to check for a person on a predetermined list such as VIPs or a law enforcement “wanted” list. The NIST evaluation team used a database of more than 26 million photos to test the software submitted by vendors. Not all of the tested algorithms performed the same and the best ones were significantly superior to the others, with accuracy being a top consideration.