When you upload your photo to one of the many social media platforms its facial recognition systems try to recognise the face and improve recognition for future uploaded content. These algorithms also take location data and people you know to create a network of data which helps in better advertising. This scale of gathering users information is a big concern for privacy.
In order to prevent companies from collecting your facial data Parham Aarabi, Professor at the University of Toronto and Avishek Bose have developed an algorithm to dynamically disrupt facial recognition systems. The technology uses a deep learning algorithm which is named as adversarial training which two AI Algorithms one-on-one which exchanges the data to improve upon themselves. Arabi and Bose are two neural networks where the first one tries to identify the faces and the second one tries to disrupt the facial data in the photo. These algorithms constantly battle each other by improving and exchanging the generated datasets.
The disrupted image is just like another Instagram filter but the algorithm is so precise that it will only affect the pixels that are not noticed by the naked eye.
“The key here was to train the two neural networks against each other — with one creating an increasingly robust facial detection system, and the other creating an ever stronger tool to disable facial detection,” says Bose, the lead author on the project. The team’s study will be published and presented at the 2018 IEEE International Workshop on Multimedia Signal Processing later this summer.
This technology is also trying to disrupt the image based search algorithms like Google, Emotion based search results.
Take your time to comment on this article.
Latest posts by Harikrishna Mekala (see all)
- A Serious Security Flaw Found in LibSSH - October 19, 2018
- Flaws in Branch.io Affected Over 685 Million Users - October 17, 2018
- Microsoft Store Has Been Hosting an Ad Clicker Disguised as a Google Photos App - October 16, 2018