Instant Face Recognition Arrives – Maybe On Your Smartphone
Sydney based start-up IDChekr claims to be developing a smartphone facial recognition app that will be able to identify people in real time by comparing their scanned image with publicly available facial images and profiles on social media and other online sources.
Founder Hayden Toppeross says IDCheckr’s secret source is recurrent neural networks. “These allow us to process extremely large data sets very quickly and refine the outputs based on usage data,” he says.
“This means the more the system is used, the more efficient it becomes. IDChekr will use open-source data sets and user photo libraries to build very large data sets.”
He says the app will be free to users. “Users will be rewarded for their input to the group data set by customisation and privacy options associated with their profile. We want this process to be enjoyable and frictionless, and above all beneficial to the user.”
IDChekr is seeking $5 million funding by November 30 through crowdsourcing Web site cootheservices.com. To date it has raised just $125,000. – Stuart Corner