Facescapes A critical guide to facial recognition technologies.

History of Facial Recognition Technology (Avery Jonas)

Manual Measurements

Mathematician, computer scientist, and founder of Panoramic Research Inc. Woodrow Wilson Bledsoe developed a manual system for classifying photos of faces through a graphical computer input device known as a RAND tablet. The RAND tablet enabled the user to input horizontal and vertical coordinates on a grid using a stylus that emitted electromagentic pulse. Bledsoe used the system to manually record the coordinates of facial features, such as eyes, hairline, mouth, and nose. The metrics could then be fed into a database so that if they user gave the system a new picture of an individual, it could retrieve an image from the database that most resembled the profile. The computer processing power limited facial recognition significantly, however, Bledsoe’s use of the RAND tablet helped proved that facial recognition was a possible viable biometric.

A Freedom of Information Act (FOIA) request and declassified documents reveal that since its inception, Panoramic had been contracted and funded by the CIA. A declassified 1968 CIA document mentions a contract with Panoramic for a facial recognition system that would reduce database search time by a hundredfold. This in the initiation of the US government’s exploration into facial recognition technology.

21 Facial Markers

There was incremental refinement of facial recognition throughout the 1970s. Harmon, Leak and Goldstein added better accuracy to manual facial recognition system through the use of 21 specific subjective markers to identify faces automatically. Markers included lip thickness and hair color.



In 1988, Sirovich and Kirby applied linear algebra to facial recognition and developed a method known as the Eigenface approach. The process began as a search for low-dimensional facial images representation. They showed that feature analysis on a collection of images could form a set of basic features.

Late 1980s to early 1990s

Automating Facial Recognition

In 1991, Turk and Pentland expanded upon the Eigenface approach by finding ways to detect faces within images. Their work was the first attempt at automatic facial recognition.


Government Programming

Defence Advanced Research Projects Agency (DARPA) and National Institute of Standards and Technology (NIST) released the Facial Recognition Technology (FERET) program to encourage the commercial facial recognition market. The project consisted of the creation of a database of 2,413 facial images, representing 856 individuals. The aim was to use the large database of test images for facial recognition, resulting in more powerful technology.

NIST also began Face Recognition Vendor Tests (FRVT) in the early 2000s. FRVTs were designed to provide independent government evaluations of facial recognition systems that were commercially available. These evaluations were designed to provide law enforcement and government agencies with information necessary to determine the best methods of facial recognition technology deployment.

1993 to 2000s

Super Bowl XXXV

Law enforcement officials began applying facial recognition in critical technology testing. The 2002 Super Bowl was a testing ground for FERET. The test was seen as a failure due to its limited success with large crowds.


Face Recognition Grand Challenge

Face Recognition Grand Challenge (FRGC) was launched in 2006 to promote and advance facial recognition technology to support existing efforts in the U.S. Government. The FRGC evaluated the latest facial recognition algorithms available, testing high-resolution face images, 3D face scans, and iris images. The results indicated that the new algorithms were much about 10 times more accurate than those of 2002. This challenge revealed the revolutionizing advancements of facial recognition technology.


Law Enforcement Forensic Database

The Sherriff’s Office in Pinellas County, Florida created a forensic database that allowed officers to tap into the photo archives of the state’s Department of Highway Safety and Motor Vehicles (DHSMV). By 2011, about 170 deputies had been allocated with cameras that let them take pictures of suspects that could be cross-checked against the database. Through this technology, more arrests and criminal investigations, which would have been impossible to obtain, were achieved.


Social Media

Social media platforms, such as Facebook, began to implement facial recognition functionalities that helped identify people whose faces may be featured in the photos that Facebook users update daily.


Neural Networks

Neural networks were introduced as the new standard in machine learning and also facial recognition. The idea which had been around for a much longer time, only then became viable due to the large amount of computer power required, as well as the large amount of data required to train a neural network.


Facial Recognition Technology in the Airport

The government of Panama partnered with the U.S. Secretary of Homeland Security at the time, Janet Napolitano, to authorize a pilot program of FaceFirst’s facial recognition platform to cut down on organized crime in Tocumen airport. The first attempt was successful as it resulted in the apprehension of multiple Interpol suspects and the technology was expanded into the north terminal facility.


Identification of Osama bin Laden

Facial recognition has been used by law enforcement and military professionals for forensics, especially in positively identifying dead bodies. Facial recognition was used to help confirm the identity of Osama bin Laden after he was killed in a U.S. raid on his compound.


Law Enforcement Adopts Mobile Facial Recognition Technology

The Automated Regional Justice Information System (ARJIS) began supplying agencies with FaceFirst’s mobile platform supporting facial recognition for law enforcement. ARJIS promotes information and data sharing among local, state and federal law enforcement agencies and through its mobile face recognition began to instantly identify people who 1) did not possess ID and 2) did not want to be identified.


iPhone X

Apple released the iPhone X, which was the first iPhone users could unlock with FaceID, Apple’s marketing term for facial recognition in the form of device security.


WatchList as a Service

FaceFirst introduced WatchList as a Service (WaaS). WaaS is a facial recognition data platform designed to help prevent shoplifting and violent crime. WatchList includes a database of known criminals that post a safety, theft or violent crime risk. The database works with the FaceFirst biometic surveillance platform to alert security about real-time threats.


Homepage Image Source: Tech Guruit