We’ve written a lot about Facial Recognition in our Market Leadership section, but we’ve never gone back to the beginnings of the technology or speculated on what the future might hold.
Biometric Facial Recognition, which was once thought to be science fiction, is quickly becoming a part of people’s daily lives.
Law enforcement, border control, retail, mobile technology, and banking and finance are just a few of the major industries that have benefited from the rapid advancements in Facial Recognition technology over the last 60 years.
As we look forward to future applications of facial recognition software, it’s important to reflect on how far we’ve come since the beginning.
Facial Recognition Makes Its Debut in the 1960s
Woody Bledsoe, Helen Chan Wolf, and Charles Bisson were among the first to develop facial recognition technology. Bledsoe, Wolf, and Bisson began working with computers to recognize human faces in 1964 and 1965.
Much of their work was never published because the project was funded by an unnamed intelligence agency. Their initial work, however, involved the manual marking of various “landmarks” on the face, such as eye centers, mouths, and so on. A computer then rotated them mathematically to compensate for pose variation. To determine identity, the distances between landmarks were automatically computed and compared between images.
The early efforts of Bledsoe, Wolf, and Bisson in Facial Recognition were severely hampered by the technology of the time, but they were an important first step in proving that Facial Recognition was a viable biometric.
Facial Recognition Accuracy Improvements in the 1970s
Following in the footsteps of Bledsoe, Goldstein, Harmon, and Lesk expanded the work in the 1970s to include 21 specific subjective markers, such as hair color and lip thickness, in order to automate the recognition.
While the accuracy improved, the measurements and locations had to be manually computed, which was time-consuming but still represents an improvement over Bledsoe’s RAND Tablet technology.
Facial Recognition Using Linear Algebra – the 1980s/90s
We didn’t see much progress with the development of Facial Recognition software as a viable biometric for businesses until the late 1980s. Sirovich and Kirby started using linear algebra to solve the problem of facial recognition in 1988.
Eigenface demonstrated how feature analysis on a collection of facial images could be used to create a set of basic features. They were also able to demonstrate that a normalized facial image could be accurately coded with fewer than one hundred values.
Turk and Pentland continued Sirovich and Kirby’s work in 1991 by discovering how to detect faces within an image, leading to the first instances of automatic facial recognition. Although technological and environmental factors hampered this significant breakthrough, it paved the way for future advances in Facial Recognition technology.
The 1990s/2000s FERET Program
The Face Recognition Technology (FERET) program was launched in the early 1990s by the Defense Advanced Research Projects Agency (DARPA) and the National Institute of Standards and Technology (NIST) to promote the commercial facial recognition market. The project entailed the creation of a facial image database. There were 2,413 still facial images in the test set, representing 856 people. The hope was that having a large database of facial recognition test images would spur innovation and lead to more powerful facial recognition technology.
Vendor Tests for Face Recognition in the 2000s
Face Recognition Vendor Tests (FRVT) were first conducted by the National Institute of Standards and Technology (NIST) in the early 2000s. Based on FERET, FRVTs was created to provide independent government evaluations of commercially available facial recognition systems as well as prototype technologies. These assessments were created to give law enforcement agencies and the US government the data they needed to figure out the best ways to use facial recognition technology.
Grand Challenge in Face Recognition – 2006
The Face Recognition Grand Challenge (FRGC) was established in 2006 with the primary goal of promoting and advancing face recognition technology to support existing face recognition efforts in the United States. Government[2].
The FRGC put the most up-to-date face recognition algorithms to the test. The tests used high-resolution face images, 3D face scans, and iris images. The new algorithms were 10 times more accurate than face recognition algorithms from 2002 and 100 times more accurate than those from 1995, demonstrating how far facial recognition technology has progressed in the last decade.
From 2010 to the present
Face recognition was first introduced to Facebook in 2010, and it helped users identify people whose faces appeared in the photos they shared on a daily basis. The feature sparked a flurry of privacy-related articles in the news media almost immediately. Facebook users, on the other hand, appeared to be unconcerned. Every day, more than 350 million photos are uploaded and tagged using face recognition, with no apparent negative impact on the website’s usage or popularity.
2017 iPhone X
From 2010 onwards, facial recognition technology advanced at a breakneck pace, and September 12, 2017 marked yet another significant milestone in the integration of facial recognition into our daily lives. This was the day Apple released the iPhone X, which was the first iPhone to feature FaceID, Apple’s marketing term for facial recognition.
Facial Recognition and NEC
Border crossings, airlines, airports, transportation hubs, stadiums, mega events, concerts, and conferences are all examples of this. Biometrics are becoming increasingly important not only in the real-time policing and security of increasingly crowded and diverse venues around the world, but also in ensuring that citizens who visit them have a pleasant experience.
NEC has developed multi-modal technologies such as face, iris, and voice recognition, finger and palmprint identification, and ear acoustic authentication, and supplemented them with AI and data analytics to enhance situational awareness and facilitate effective real-time or post-event action in both law enforcement and consumer-oriented spheres, as a long-time committed pioneer of biometric research and solutions.
Facial Recognition Technology’s Future
As we approach 2020, facial recognition technology continues to advance at a rapid pace, and its applications are becoming more widespread. In a recent post, we discussed the eight Facial Recognition trends to watch in 2020. These included the following:
- Retail
- Resorts and Hospitality
- ATMs
- Advertising on the Internet
- Bus Security
- Airlines
- Customer Experience that is Tailored
- Stores with no employees
Make sure to read our posts to learn how each of the industries listed above is embracing and putting facial recognition technology to good use.
When used correctly, facial recognition technology has the potential to transform our lives and the way we interact with the world.