emotion detection


Natural Human-Drone Interaction

Research project from Eirini Malliaraki illustrates ideas for drone programming, from gesture to emotion recognition:

1-month graduate project // Royal College of Art & Imperial College// May 2017

Taking inspiration from the interaction between falconers and their birds of prey, as well as from common daily gestures, cybernetics, dance, and robotics, several themes were explored, namely:
- a gesture-based interaction scheme that attempts to create a more intuitive and natural way to communicate with aerial robots
- ways in which aerial robots can become more autonomous by interpreting their environment in richer ways
- ways in which they can communicate their intentions and give feedback
- ways in which an aerial robot can understand and react to human emotions and eventually influence our behaviour

Parrot AR Drone, Node js, Javascript, Affectiva Emotion analysis SDK


Someone needs to look after Dirk, okay?

Like, he wakes up in the morning, probably on Todd’s sofa, all rumpled and uncomfortable from sleeping there. And he finds that someone put a blanket on him last night, which was nice of them, and he smiles a little. Pulls it around his shoulders. And then Todd emerges from the kitchen, because this is the one time he’s ever been up before Dirk (because it had been a tough case they just solved, so he figured Dirk could use the rest) and Todd makes him a cup of tea. And they sit and talk and Dirk’s all wrapped up in his blanket with his tea and he has this smile on his face, and says,
“Thanks Todd.” In that little way he does.

The giraffe man just needs a BREAK, okay?


12 days of stydia
    ⤷  day two:
favorite episode (3x11)

Brotzly Week Day #4: ‘Laugh’

Look, Todd knew he could be a bit of a miserable asshole. It was true. He didn’t smile all that much (although that had been changing recently, and he suspected a certain holistic detective was to blame), and he laughed even less. It was a fact of his life, and he was at peace with it.

Dirk, however, was not.

Todd was so caught up in a rare belly-laugh at one of Amanda’s jokes, that he didn’t notice the quiet beep of Dirk’s phone right as he started. He noticed the one after the laugh was done though, and he turned his head to find Dirk pointing his phone directly at Todd, face stretched in a triumphant beam.

“Got it!” said Dirk proudly, withdrawing the phone and flicking the screen.

“What? Got what, what’s going-?”

He was cut off by the sound of his own laughter ringing out from Dirk’s speakers. Dirk smiled smugly, tapping the screen a few more times. “Looks like I’ve got a new ringtone!”

“What the-? Dirk, what the hell?!”

“Calm down, Todd,” said Dirk with a light snort. “You’ve got a lovely laugh and I don’t hear enough of it! I just thought I’d take matters into my own hands.”

“Dirk, you can’t have me laughing as your ringtone it’s… it’s weird!”

“People have their significant others as lockscreens, no?” Dirk argued, turning his phone for Todd to see. Sure enough, the screen was filled with an unflattering picture of Todd’s sleeping face pressed into the couch cushions. “I think it completes the theme.”

“Dirk, it’s creepy!” Todd insisted, flustered. “People are gonna hear that and, and- how would you like it, huh? If I just- just used your voice as a text alert without permission? It’d be weird, right?”

Dirk looked… a little too happy. “I’d be touched, actually.”

“Oh, yeah? How ‘bout we put that theory to the test, hm?” Todd said, sliding his own phone out of his pocket and flicking to settings (completely ignoring the photo of Dirk attempting to play Todd’s guitar with his tongue sticking out in concentration on the lock screen- it wouldn’t help his case right now). “Just a sec- there!” he tapped ‘preview’ and Dirk’s laugh rang tinnily out of the speakers, filling the office. “Not so flattering now, right? Anyone within ten feet of me is gonna hear that every time I get a message and-“

It was at that moment that he remembered who else currently was within ten feet of him, and he stiffened. Slowly, he turned his head back round, and came face to face with his sister’s look of delighted disbelief.

“So,” she said with a mischievous twinkle in her eye. “You’ve already got his laugh recorded, huh?”

Shit. “I… I was trying to prove a point!”

“What point?”

That Dirk shouldn’t be self-conscious about his laugh because it’s one of the cutest I’ve ever heard. “That… that he sounds like a donkey when he laughs.”

“Oh, really?”

Dirk looked at Todd, eyes wide. “So… you really don’t like it?”

SHIT. “Wha- no! No, Dirk, I love it, really, I-!”

“Ha!” Amanda yelled, triumphantly finger-gunning. “Knew it! You’re a fucking sap!”

“No, I’m not, I’m-!”

“Ah ah ah, too late,” she crowed, slapping him on the back. “Dude, it’s okay. You’re a grown-ass man with a cute dorky boyfriend- you can quit pretending to be edgy!”

Todd wanted the ground to swallow him up. “Hey, I’m not pretending anythi-!”

“Uh huh, yeah, tell it to someone who still buys your bullshit,” she said, shaking her head and standing to shrug on her jacket. “Sorry, guys. Hate to break it to you but you’re fucking cute. Kudos. Later, nerds. Enjoy your matching ringtones!”

She sauntered off, probably to go bug Farah instead, giving Dirk the customary wave and Todd the customary middle-finger on her way out.

Todd glowered after her, and turned to Dirk- who had far too smug a look on his face. “We’re not- I’m- I am not… cute!”

“Whatever keeps you sane,” Dirk simpered, eyes crinkled with amusement. “What’s for dinner?”

“…I was gonna make spaghetti.”

“Sounds lovely, dear.”

“Shut the fuck up.”


Human Pose Detection - Mining Body Language from Videos

Machine Learning investigation from samim examines body language in video using recently released open-source code library OpenPose:

From Gene Kelly’s Step-Dance to Bruce Lee’s Kung-Fu — iconic movement has made history. Communicating through Body Language is an ancient art form, currently evolving in fascinating ways: Computationally detecting human body language is becoming effective and accessible. This experiment explores enabling technologies, applications & implications.

For over 20 years, Motion Capture has enabled us to record actions of humans and then use that information to animate a digital character or analyse poses. While movie makers and game developers embraced such technologies — it until recently required expensive equipment which captured only few aspects of the overall performance.

Today, a new generation of machine learning based systems is making it possible to detect human body language directly from images. A growing number of research papers and open-source libraries addresses key aspects: Body, Hand, Face, Gaze Tracking. Identity, Gender, Age, Emotion and Muscle strain Detection. Action Classification & Prediction. We now can…

More Here