Though security using face recognition solutions are believed to be the wave of the future, it has increasingly sparked controversy. For instance, consider a teenager suing a top tech company for $1 billion for the alleged use of facial recognition that led to him being arrested for thefts he didn’t commit. This is the aftermath of a face detection solution’s technical glitch which mistakenly linked the student with the face of the real thief. Whatever the truth turns out to be, this lawsuit is the latest sign of an escalating backlash against facial recognition failures. However, none of AnyVision’s client has ever experienced such a dramatic scenario. A designer and developer for face recognition platform, AnyVision was founded in 2014 to solve real-world security problems—customizable and reliable face detection. AnyVision leverages best-in-class facial biometric technology, demonstrates meaningful ROI to clients across industries, and assembles a strong team. The company has been awarded for both one to one and one to many face recognition by leading academic benchmarks.

At the cockpit of AnyVision is Eylon Etshtein, a mobile and internet entrepreneur and the former owner of Puzzle Interactive—one of the first digital companies in Israel. Etshtein eats and breathes technology; designs, create and has knowledge in programming almost in any language in the digital world. Being an expert in thinking outside-the-box and briefing complex projects, it is Etshtein’s unrelenting passion for security solutions that are spearheading AnyVision to disrupt the facial biometrics solutions landscape.

The Deeper Nodes of AnyVision

Unlike standard solutions in the market, AnyVision’s software technology learns from the video feed, is customizable, and runs on multiple platforms including any existing CCTV infrastructure. Just integrate AnyVision and empower the sensors with superpower and verifying the identity of an individual is not futurist anymore. When unfamiliar faces are captured in the CCTV cameras, the solution trigger alarms and send instant alerts. AnyVision's advanced facial biometric technology’s flexibility and plug-n-play capabilities blow other solutions out of the water.

The company’s facial recognition process has four stages: detection, feature extraction, tracking, and recognition. For the detection process, AnyVision’s AI platform divides the frame into a grid and starts searching for a face. Once the face is detected, a mathematical model is extracted. This process is then repeated in the next frame, creating a track. As the track is completed, the algorithm creates a combined model based on all the frames it has analyzed.

This allows detecting people in extreme conditions such as high contrast, profile view, low resolution, and even disguise subject. The platform will use this model to search for any particular database, and if it finds a match, an alert will be triggered automatically. This process happens simultaneously within a multi-camera environment on all phases in the field of view. AnyVision also champions its technology’s scalability: “It’s not just how accurate the system is, it’s also how much it scales,” says Eylon Etshtein, Co-Founder and CEO, AnyVision. “It provides the capability to add more than 20 concurrent full HD camera streams on a single GPU.”
An innovative offering from AnyVision, Better Tomorrow—a mobile app that provides fast and accurate facial recognition capability for field operatives. Better Tomorrow will detect any face in the field of view of the camera and create an alert when it finds a match to the suspect's list. The app works with off the shelf smartphones and can either utilize the phones camera or stream videos from anybody worn camera. It will then use the smartphones chip to process the footage. The application also allows the agent to edit the suspect's list, track the location of each detection, and work with an offline database of up to ten thousand suspects.

Meanwhile, the company’s SesaMe SDK provides a seamless and secure way for users to involve and authenticate into their account. The enrolment process is easy. Using the smartphone camera, the user takes a snap of his ID and then chooses a quick video. SesaMe will verify that the ID is legitimate and use the video to make sure that the user’s identity is legitimate. Then using the cutting edge facial recognition technology, SesaMe will compare between the two to see that the identification data match and verify with the database to make sure that the particular user doesn’t already exist. This entire process takes only a few seconds, and authentication is more effortless. The user just has to look into the camera, and that’s it.

A Blend of Simplicity and Innovation

At AnyVision, the team believes in keeping things simple. They consider that technology should serve by giving the right information exactly when needed. AnyVision’s face recognition solution is the result of over two decades of academic research and field experience. Behind the scenes, it is AnyVision’s highly talented and certified team of specialists who give an edge to the company. AI designers and security veterans combined, AnyVision team’s sophisticated face recognition platform is used across multiple industries to protect citizen’s privacy by exposing only the identities of parties of interest.

Since its inception, AnyVision has worked with several partners including Microsoft, Google, Telefonica, Axis Communications and Johnson Controls in enhancing their facial recognition capabilities and finding suspects of critical crimes.

Leading technology partners like Genete allow AnyVision to utilize their cutting-edge technology and offer an effective and complete security solution.

AnyVision recently joined NVIDIA’s Metropolis Software Partner Program (MSPP) to combine the company’s robust face and human recognition platform with the power of NVIDIA GPUs. By leveraging NVIDIA technology, AnyVision can turn any existing CCTV infrastructure in cities worldwide into a network of smart cameras with the capability to offer highly accurate real-time actionable insights.

These applications can be applied across a wide variety of verticals to track, detect, and recognize any person of interest in a large crowd with more than 99 percent accuracy. Furthermore, AnyVision’s technology is proven scalable across all platforms, while maintaining an optimal level of personal data protection. “At AnyVision, our goal is to create a safer and more secure environment for a better tomorrow. By working with NVIDIA and applying its technology enables us to employ our AI algorithms into our software, we will sustain a safe, accurate, and optimized system that provides effective security measures,” says Etshtein.

"AnyVision provides the capability to add more than 20 concurrent full HD camera streams on a single GPU"

The company recently partnered with Moscow’s Domodedovo Airport Authority and is testing the compatibility of AnyVision’s facial recognition solution to be used by the airport security and law enforcement agencies. The system was installed at one of the airport’s entrances, and a photo database of airport employees was compiled for testing purposes. In December it was revealed that facial recognition is operational on Moscow’s municipal CCTV surveillance network and that the software can return weeks worth of video footage in 10 seconds for one individual among 10 million faces stored in the system.

Redefining Machine Vision

Headquartered in Tel Aviv, AnyVision has more than 200 employees worldwide, with offices in Mexico, New York, London, Singapore, and Belfast—a location alone that houses a research team of over 30 PhD holders.

The company has already expanded from 20 to 130 employees in the past two years with offices opened in the US and UK and plans to spread its wings to Los Angeles, London, and Singapore. Etshtein expects 2019 to be a year in which AnyVision sustains its most loyal customers—the ones who have been using facial biometric solutions without interruption—and win over new buyers. With growing global and industry demand, AnyVision is also expanding its offerings and capabilities to offer AI solutions to HLS/Police, airports, sports/entertainment, smart cities, critical infrastructure, banks, and retail verticals.