Skip to main content

Researchers Design a Filter to Defeat Facial Recognition Software

Most social networks these days, including Facebook, use facial recognition on your photos whenever you make an upload to gather data on you. In addition to social networks, the web is also full of similar software that is able to automatically recognize a person and figure out other things such as ethnicity, gender, and the environment.
As privacy concerns have grown recently, University of Toronto researchers have developed an AI that counters facial recognition software in a subtle way, by making minute adjustments to your photos. The research, led by professor Parham Aarabi and Avishek Bose (an MASc candidate), has successfully disrupted common facial recognition systems.
Personal privacy is a real issue as facial recognition becomes better and better. This is one way in which beneficial anti-facial-recognition systems can combat that ability.

ALSO, READ

AI vs AI

In order to strengthen their solution, the team used an AI technique known as adversarial training in which two opposing neural networks consistently learn from each other. So in their case, they developed two AIs: one that recognizes faces, and a second one that disrupts it by constantly training itself to counter the first one.
This way, both AI’s keep learning from each other and keep expanding on ways to prevent facial recognition. Avishek Bose says,

ALSO, READ

The disruptive AI can ‘attack’ what the neural net for the face detection is looking for. If the detection AI is looking for the corner of the eyes, for example, it adjusts the corner of the eyes so they’re less noticeable. It creates very subtle disturbances in the photo, but to the detector they’re significant enough to fool the system.
We can expect it work like a filter or a third-party plug-in that slightly modifies your image for you before you upload it to a social network. The researchers tested it out using a library of 600 different faces that were originally detectable, the new solution managed to reduce facial detection accuracy from 100 to 0.5 percent.
Via UoT

Comments

Popular posts from this blog

POS

POINT OF SALE Point of sale  (also called  POS  or  checkout , during computerization later becoming  electronic point of sale  or  EPOS ) is the place where a retail transaction is completed. It is the point at which a customer makes a payment to the merchant in exchange for goods or services. At the point of sale the retailer would calculate the amount owed by the customer and provide options for the customer to make payment. The merchant will also normally issue a receipt for the transaction. The POS in various retail industries uses customized hardware and software as per their requirements. Retailers may utilize weighing scales, scanners, electronic and manual cash registers, EFTPOS terminals, touch screens and any other wide variety of hardware and software available for use with POS. For example, a grocery or candy store uses a scale at the point of sale, while bars and restaurants use software to customize the item or service s...

This startup is pairing the blockchain with farming

Ripe, a blockchain startup, is  showcasing  how the distributed ledger technology can be applied to farming and provide companies with important data on crops and throughout the supply chain. By combining a blockchain with purpose-built scanners and specialized sensors, the company is able to provide critical data that is easily communicated to partners, demonstrating why  blockchain is growing more popular  as a tool to be leveraged as part of an IoT solution. Blockchain is a key tool for the IoT because of how it combines data storage and communication in a trusted medium. A blockchain is a shared, distributed database  that acts as an immutable ledger, recording entries and verifying them across a number of independent participants. Once an entry is made, it’s marked with a unique hash code that places it sequentially in the ledger. In agriculture, this means that farmers can use sensors and handheld devices to gather data about crops like tomatoes...

Li-Fi

Li-Fi Light Fidelity  ( Li-Fi ) is a bidirectional, high-speed and fully networked wireless communication technology similar to Wi-Fi. The term was coined by Harald Haas and is a form of visible light communication and a subset of optical wireless communications (OWC) and could be a complement to RF communication (Wi-Fi or cellular networks), or even a replacement in contexts of data broadcasting. It is wire and uv visible-light communication or infrared and near-ultraviolet instead of radio-frequency spectrum, part of optical wireless communications technology, which carries much more information and has been proposed as a solution to the RF-bandwidth limitations. Technology details This OWC technology uses light from light-emitting diodes (LEDs) as a medium to deliver networked, mobile, high-speed communication in a similar manner to Wi-Fi. The ...