The neural network was trained to copy movements of people from videos on YouTube

Date:

2018-10-11 17:45:07

Views:

28

Rating:

1Like 0Dislike

Share:

The neural network was trained to copy movements of people from videos on YouTube

Today, the main methods of creating realistic animation of virtual creatures are skeletal animation and the infamous Motion Capture (mocap). However, experts are constantly in search of new lower quality, but cheaper ways. And, quite possibly, this may help the neural network. One of them recently trained to recognize the movements of people and transfer them to virtual characters. Perfect for any video. Even a YouTube video.

For the development of responsible scientists from the Laboratory of research in artificial intelligence at the University of California at Berkeley (USA) and the University of British Columbia (Canada). The new algorithm is called DeepMimic was created in early 2018, but the full presentation was held only recently at the conference SIGGRAPH Asia in 2018. According to the authors,

"Our approach is based on a thorough analysis and evaluation of certain poses and events. Neural networks with deep machine learning can perform qualitative analysis of any movements from publicly available sources such as YouTube. It has a huge potential for the development of tools for a quick and simple animation, just by requesting the video with the desired behavior on the Internet. Our algorithm is able to explore a wide range of skills including acrobatics and martial arts."

The algorithm works is quite simple: the system "looks" video, which is performed a certain action. Then the movie is broken down frame by frame and for each frame created a three-dimensional model. After that all models are "connected", and a special subroutine monitors the ligaments between them and removes artifacts and smoothes out the "choppy" animation.

To Evaluate the performance of the algorithm, you can demo video is available below. But I want to add that the result was really impressive, provided that it is not, of course, pre-prepared videos, and random videos from the Internet.

To Discuss new technology to create animation and other news, you can in our

Recommended

Date of first launch rocket

Date of first launch rocket "Soyuz" after the accident

according to the Telegram channel "house of cards: Russia", citing unnamed sources, the first launch of the Soyuz rocket after an accident is scheduled for October 25-26. We will remind, on October 11, immediately after the start of the rocket "Soyuz...

Blue phosphorus is a new two-dimensional material capable oust graphene

Blue phosphorus is a new two-dimensional material capable oust graphene

Invented in the early 21st century, graphene has already found its application in many fields of science and technology. And even gave the scientists who studied him the Nobel prize. However, two-dimensional structure like that of carbonaceous materi...

SoundBender: the new technology of ultrasonic levitation

SoundBender: the new technology of ultrasonic levitation

for many years scientists from all over the world are trying to develop an efficient levitation technology, because it will open the way not only to create new vehicles (Hello, a hoverboard!), but that would change our whole surrounding space. There ...

Comments (0)

This article has no comment, be the first!

Add comment

Related News

Almost all graphene available on the commercial market, was a fake

Almost all graphene available on the commercial market, was a fake

Despite all the hype , all of its properties and promises of scientists, you may be surprised by the fact that this material is still not used everywhere. As it turned out, this is not surprising. An international group of scienti...

Masternode simple language. How to choose masternode and how much it can earn

Masternode simple language. How to choose masternode and how much it can earn

In September in the world of cryptocurrencies the new coin — ANON. It was a fork of Bitcoin and Zclassic and promises users an unprecedented level of privacy. But another surprise: at first, masternode ANON showed an 1800% ROI. It...

Created material, 1 gram of which

Created material, 1 gram of which "fit the field"

Imagine a football field. In the background you'll seem short. Now, try to imagine that this enormous size of a flat football field to fit on a smaller area. What is the minimum size will be able to achieve? You will be surprised...