- TikTok star Mohit Mor shot and killed 7 Months Ago
- Stephen A. Smith is baby 7 Months Ago
- Tfue releases statement on FaZe Clan lawsuit, says his contract is ‘f*cked’ 7 Months Ago
- People are using an app to out gropers on Japan’s subway Today 11:24 AM
- Trump misspelled ‘accomplishments’ on handwritten notes, photo shows Today 11:12 AM
- HUD proposal would allow homeless shelters to refuse trans people Today 10:44 AM
- Disney’s ‘Aladdin’ remake isn’t terrible Today 10:11 AM
- Police under investigation after running over 1-year-old child Today 9:16 AM
- Who is Jannah, the breakout star of ‘Star Wars: The Rise of Skywalker’? Today 9:10 AM
- Trump revives his ‘dumb as a rock’ insult for Rex Tillerson Today 9:03 AM
- Forget Hot Jafar. All hail Fat Ursula Today 8:14 AM
- How to watch ‘The Affair’ for free Today 8:00 AM
- Olivia Wilde’s ‘Booksmart’ is a teen comedy that will actually age well Today 7:45 AM
- Conservative rising star Kyle Kashuv busted using the N-word a whole bunch Today 7:14 AM
- How to watch ‘The Name of the Rose’ for free Today 7:00 AM
One small step for man, a galloping dance craze for robot-kind.
In a new paper to be published next month in the Association for Computing Machinery’s Transactions on Graphics journal, researchers from the University of California-Berkeley were able to train a deep neural network to copy human movements by simply feeding them YouTube videos, paving the way for better mimicry of people.
Humanoid characters on a computer simulation were able to do backflips, handsprings, and cartwheels after learning them from video clips through state-of-the-art techniques in computer vision and reinforcement learning.
Here’s one example.
Xue Bin Peng and Angjoo Kanazawa, two of the artificial intelligence researchers who developed the program, said in a blog that this is a departure from previous techniques which strongly restricted the behaviors which can be produced.
“Therefore, these methods tend to be limited in the types of skills that can be learned, and the resulting motions can look fairly unnatural. More recently, deep learning techniques have demonstrated promising results for visual imitation on domains such as Atari and fairly simple robotics tasks,” the pair said.
According to them, their system works by first predicting the pose of the subject of a video fed in the pose estimation stage. After this, the motion reconstruction stage collates pose predictions into a reference motion and fixes artifacts for smoother movement. Finally, the reference motion is passed to the motion imitation stage, where a character is trained to mimic motion using reinforcement learning.
Researchers were able to teach simulated characters more than 20 different skills like vaulting, jumping jacks, high kicks, pushing a box, dancing from side to side, running, and walking.
“The key is in decomposing the problem into more manageable components, picking the right methods for those components, and integrating them together effectively. However, imitating skills from videos is still an extremely challenging problem, and there are plenty of video clips that we are not yet able to reproduce,” Xue and Kanazawa said.
Unfortunately, one of the actions they cannot properly reproduce yet is the 2012 viral dance hit “Gangnam Style.”
“We still have all of our work ahead of us, and we hope that this work will help inspire future techniques that will enable agents to take advantage of the massive volume of publicly available video data to acquire a truly staggering array of skills,” Xue and Kanazawa said.