3 minute read

Face Recognition

"When your face will become your password, unlocking smartphones and bank accounts, From Facebook’s Deep view From Amazon to Apple Face Recognition in 2018 will change our lives"

Advertisement

Did you ever sit back and wonder how your phone detects your face in a photo or how Facebook automatically identifies you and your friends when you upload your memories?

The technology, formally known as Facial Recognition and Detection is an emerging trend in the Machine Learning and Artificial Intelligence segment. It identifies and verifies the identity of a person based on various facial parameters like skin, the shape of your eyes and lips. The technology was brought into the limelight when Apple Inc launched Face-ID as a special feature in the iPhone X. Earlier smartphones which supported face detection wasn’t as advanced as the current version.

Many of the interesting advantages of this biometric system are its non-invasive or contactless nature, cheaper than other access-restriction systems and can be used for a large number of people at once.

HOW DOES IT WORK?

This technology needs a set of parameters and a database to store and compare the data collected. It also needs a strong algorithm to analyse and give out results in real-time. Python combined with Machine Learning (ML) is used to implement these requirements.

The algorithm matches your face with billions of other faces in the database based on key factors like distance between your eyes and the shape of your chin and generates a unique numerical number to store this data.

• Techniques used are skin texture analysis where the texture of a patch of the skin is stored as numbers.

• 3-D algorithms use features like eye sockets, jawlines, and skull shape too

• Thermal algorithms just store the shape of a certain individual.

Modern efficient computers combine all these algorithms to ensure accuracy.WHERE CAN IT BE APPLIED?

Security: The most widely used and researched feature, is to prevent and detect crime almost immediately. Airports and famous public places continuously scan the area for any kind of suspicious individuals being searched by law agencies around the world. This includes terrorists, anti-nationals and other threats to humanity. The industry is expected to grow from $4 billion to $8 billion within 4- 5 years and be the most extensively used security feature.

o Phone access. o Helping law enforcement by identifying people from a safe distance using drones o Forensic Investigations o Preventing School Threats – Identifying sex offenders, drug dealers or violent parents. o Air Travel Is More Convenient – Long queues to verify documents and people can be reduced while checking in.

Marketing: Facial recognition can guess or predict the age and gender of a human being and tinker advertisements and infomercials to entertain them.

“I think it’s only a matter of time before this technology revolutionizes retail shopping. It presents the opportunity for retailers and customers to draft a new contract, one that protects privacy while also delivering a far more seamless and personalized shopping experience.”

-Peter Trepp, CEO, FaceFirst

❖ Smarter Advertising ❖ Identify people on Social Media Platforms- Facebook’s DeepFace is an apt example. ❖ Recognize VIPs in a crowd

Facilitate Secure Transactions – Cards removed the need of carrying cash, but now you don’t need to carry anything, as the tech identifies you using a special scanner and pays for your purchases.

MISC:

Track Attendance in universities and offices easily Government Services Find missing people from CCTV footage around the world. Facial Recognition can also be used to track pets and ensure their safety.

PROS AND CONS

Facial Recognition is cheaper, faster than fingerprint readers and can be used in almost all work environments, using infrared or normal light waves. It is contactless and ensures time-efficiency.

In the privacy-fearing era, people accessing personal digital data by just having your face is a scary thought. There is no liability with respect to the data being used for illicit activities like identity theft. The process is not yet 100% accurate and can be fooled by people having selfish aims.

In conclusion, the community should be taking initiative to ensure the smooth transition of modern technology while maintaining the balance between basic freedoms.

“Sure, there are valid concerns about intrusion and privacy regulations, but follow the rules (such as full disclosure, opt-in and appropriately handled personal data storage) and the results will be a transparent system which only succeeds if it provides value to the individuals concerned.”

-Drew Bates, Head of Product Marketing, SAP Innovation Lab