Facebook's Facial Recognition Software Still Has A Ways To Go Social Media Insights is a daily newsletter from Business Insider Intelligence that collects and delivers the top social media news first thing every morning. You can sign up to receive Social Media Insights here or at the bottom of this post.
Facial recognition software will help Facebook organize the billions of photos on its platform, but in the wake of the NSA scandal, many users are wondering how such technology might be used by law enforcement. Facebook has the largest human identification system in the world, "and it's all been formed by people voluntarily submitting pictures to Facebook and identifying who they belong to," said Amie Stepanovich, director of the domestic surveillance project at the Electronic Privacy Information Center. However, Facebook's facial recognition software still has a long way to go before it can accurately identify someone out of a crowd of people. What's more, surveillance images are often low-quality, and that influences the accuracy of facial recognition software. (NPR) Google is reportedly building a "secret" floating data center in the SF Bay. (CNET) Snapchat is in talks to raise a new round of funding at a $3.6 billion valuation. (AllThingsD)  In a new report from BI Intelligence, we break down the demographics of Facebook and show why it's such a powerful platform for companies, brands, and products. Facebook's surprisingly strong statistics in terms of gender breakdown, income levels, and age diversity makes it the obvious go-to marketing platform. Our report also spotlights similar statistics for all the main social networks, but Facebook's numbers are perhaps the most compelling. (BI Intelligence)
Michael Arrington thinks Google should buy Twitter before its IPO. (Uncrunched) LinkedIn addressed security concerns about its new app, Intro, which has access to users' email. (LinkedIn Blog) Pinterest closed a deal with Getty Images to ensure that photos are properly attributed on the popular photo-sharing site. (TechCrunch) | 
Facebook
Twitter
Digg
Reddit
StumbleUpon
LinkedIn
No comments:
Post a Comment