CLOSE
George Takei Presents on YouTube // Microsoft's Project Oxford
George Takei Presents on YouTube // Microsoft's Project Oxford

Microsoft's Emotion Recognition Software Quantifies Feelings Based on Expressions

George Takei Presents on YouTube // Microsoft's Project Oxford
George Takei Presents on YouTube // Microsoft's Project Oxford

Can you tell how someone is feeling just by looking at their face? Microsoft's Project Oxford team has developed an experimental tool that it says can detect a person's emotional state just by analyzing a photo of his/her expression. Each emotion is given a decimal value to five places, which together equal .99999 and form a numerical representation of complex human feelings.

The emotions that the API claims to recognize are anger, contempt, disgust, fear, happiness, neutral, sadness, and surprise. "These emotions are understood to be cross-culturally and universally communicated with particular facial expressions," reads the MPO website. The tool can recognize and analyze as many as 64 faces in a single image, but it does include the disclaimers that  "frontal and near-frontal faces have the best results" and "recognition is experimental, and not always accurate."

Hyperallergic tested the tool to see how it analyzed the emotions of famous works of art. It found that Mona Lisa's smirk was half happiness and half neutral, and it attributed the expression of Dorothea Lange's Migrant Mother (1936) as being mostly neutral with less sadness than the iconic photo actually conveys. Using photos from iStock, YouTube, and Wikimedia Commons, we decided to try the tool for ourselves to see how accurate or inaccurate it could be (note: it does not work on photos of angry cats). Check out the results below, and conduct your own emotional photo experiment on Microsoft's Project Oxford website.


Puddles Pity Party on YouTube // Microsoft's Project Oxford


[h/t Hyperallergic]

nextArticle.image_alt|e
iStock
arrow
Big Questions
What Could the Repeal of Net Neutrality Mean for Internet Users?
iStock
iStock

What could the repeal of net neutrality mean for the average American internet user?

Zouhair Belkoura:

The imminent repeal of net neutrality could have implications for Americans beyond the Internet’s stratification, increased costs to consumers, and hindered access to content for all. Net neutrality’s repeal is a threat to the Internet’s democracy—the greatest information equalizer of our time.

With net neutrality’s repeal, ISPs could be selective about the content and pricing packages they make available. Portugal is a good example of what a country looks like without net neutrality

What people may not realize is that a repeal of net neutrality would also give ISPs the ability to throttle people’s Internet traffic. Customers won’t likely have visibility into what traffic is being throttled, and it could substantially slow down people’s Internet connections.

What happens when this type of friction is introduced to the system? The Internet—the greatest collective trove of information in the world—could gradually be starved. People who experience slower Internet speeds may get frustrated and stop seeking out their favorite sites. People may also lose the ability to make choices about the content they want to see and the knowledge they seek.

Inflated pricing, less access to knowledge, and slower connections aren’t the only impact a net neutrality repeal might have. People’s personal privacy and corporations’ security may suffer, too. Many people use virtual private networks to protect their privacy. VPNs keep people’s Internet browsing activities invisible to their ISPs and others who may track them. They also help them obscure their location and encrypt online transactions to keep personal data secure. When people have the privacy that VPNs afford, they can access information freely without worrying about being watched, judged, or having their browsing activity bought and sold by third-party advertisers.

Virtual private networks are also a vital tool for businesses that want to keep their company data private and secure. Employees are often required by their employers to connect to a VPN whenever they are offsite and working remotely.

Even the best VPNs can slow down individuals' Internet connections, because they create an encrypted tunnel to protect and secure personal data. If people want to protect their personal privacy or company’s security with a VPN [they] also must contend with ISP throttling; it’s conceivable that net neutrality’s repeal could undermine people’s freedom to protect their online safety. It could also render the protection a VPN offers to individuals and companies obsolete.

Speed has always been a defining characteristic of the Internet’s accessibility and its power. Net neutrality’s repeal promises to subvert this trait. It would compromise both people's and companies’ ability to secure their personal data and keep their browsing and purchasing activities private. When people don’t have privacy, they can’t feel safe. When they don’t feel safe, they can’t live freely. That’s not a world anyone, let alone Americans, want to live in.

This post originally appeared on Quora. Click here to view.

nextArticle.image_alt|e
Andreas Rentz/Getty Images
arrow
Pop Culture
An AI Program Wrote Harry Potter Fan Fiction—and the Results Are Hilarious
Andreas Rentz/Getty Images
Andreas Rentz/Getty Images

“The castle ground snarled with a wave of magically magnified wind.”

So begins the 13th chapter of the latest Harry Potter installment, a text called Harry Potter and the Portrait of What Looked Like a Large Pile of Ash. OK, so it’s not a J.K. Rowling original—it was written by artificial intelligence. As The Verge explains, the computer-science whizzes at Botnik Studios created this three-page work of fan fiction after training an algorithm on the text of all seven Harry Potter books.

The short chapter was made with the help of a predictive text algorithm designed to churn out phrases similar in style and content to what you’d find in one of the Harry Potter novels it "read." The story isn’t totally nonsensical, though. Twenty human editors chose which AI-generated suggestions to put into the chapter, wrangling the predictive text into a linear(ish) tale.

While magnified wind doesn’t seem so crazy for the Harry Potter universe, the text immediately takes a turn for the absurd after that first sentence. Ron starts doing a “frenzied tap dance,” and then he eats Hermione’s family. And that’s just on the first page. Harry and his friends spy on Death Eaters and tussle with Voldemort—all very spot-on Rowling plot points—but then Harry dips Hermione in hot sauce, and “several long pumpkins” fall out of Professor McGonagall.

Some parts are far more simplistic than Rowling would write them, but aren’t exactly wrong with regards to the Harry Potter universe. Like: “Magic: it was something Harry Potter thought was very good.” Indeed he does!

It ends with another bit of prose that’s not exactly Rowling’s style, but it’s certainly an accurate analysis of the main current that runs throughout all the Harry Potter books. It reads: “‘I’m Harry Potter,’ Harry began yelling. ‘The dark arts better be worried, oh boy!’”

Harry Potter isn’t the only work of fiction that Jamie Brew—a former head writer for ClickHole and the creator of Botnik’s predictive keyboard—and other Botnik writers have turned their attention to. Botnik has previously created AI-generated scripts for TV shows like The X-Files and Scrubs, among other ridiculous machine-written parodies.

To delve into all the magical fiction that Botnik users have dreamed up, follow the studio on Twitter.

[h/t The Verge]

SECTIONS

arrow
LIVE SMARTER
More from mental floss studios