Pranav Dar — March 7, 2018
AVbytes

Overview

  • A machine learning algorithm can simulate the way animals walk
  • Developed by researchers at the University of Manchester, it maps the movement incredibly accurately
  • The algorithm uses the movement of the animal to generate a skeleton model and skin outline
  • Check out how the algo works, along with the video below

 

Introduction

From the rhinoceros in Black Panther to Gollum in Lord of the Rings, humans have always had a fascinating with computer simulated animals. The results are usually varied – but we still haven’t got to a point where they look exactly lifelike.

That might be about to change.

Researchers at the University of Manchester have developed and fine-tuned a machine learning algorithm that improves the computer simulations of animals.

A report on phys.org revealed that the animal being used behind the machine learning model was a chimpanzee. This will not only help researchers improve their understanding of how the locomotion of animals functions, but also improve the technology behind it.

                                                                       Credit: Bill Sellers

How was the model developed?

The idea behind the algorithm was to calculate the energy it takes to walk in a stable fashion, as compared to other movement patterns. A full-body scan of the chimpanzee was taken to create the model. Using this scan, the researchers generated an outline of the skeleton and the skin.

The skeletal model was then used to define the joints, muscles and limbs for the simulation. The final step involved analyzing the movement of the animal while it was made to walk.

Check out the video of the simulated chimpanzee walking below:

 

Our take on this

This study will be of great help to the animal bio-mechanical research community. Until now, machine learning algorithms were on the basic scale when it came to replicating the movement of animals. This should help speed along the process for everyone.

But outside that field, can you imagine how much more realistic those simulated animals will become in movies? From animated features to novel adaptations, this could be a potential game changer once it arrives in mainstream cinema.

 

Subscribe to AVBytes here to get regular data science, machine learning and AI updates in your inbox!

 

About the Author

Pranav Dar

Senior Editor at Analytics Vidhya. Data visualization practitioner who loves reading and delving deeper into the data science and machine learning arts. Always looking for new ways to improve processes using ML and AI.

Our Top Authors

  • Analytics Vidhya
  • Guest Blog
  • Tavish Srivastava
  • Aishwarya Singh
  • Ram Dewani
  • Faizan Shaikh
  • Aniruddha Bhandari

Download Analytics Vidhya App for the Latest blog/Article

3 thoughts on "Machine Learning is Helping Researchers Create Lifelike Animal Simulations"

MahtabMd
MahtabMd says: March 08, 2018 at 5:10 pm
Why just in movies. They will create animal robots with those movements. Reply
Pranav Dar
Pranav Dar says: March 08, 2018 at 5:13 pm
Absolutely. It's already being done to stunning effect by companies like Boston Dynamics but with this level of granularity, we might not even be able to tell the difference between real and AI in a decade's time. Reply
Jean-Claude KOUASSI
Jean-Claude KOUASSI says: March 13, 2018 at 3:34 pm
That is particularly interesting. Perhaps, the results of these researches will help improve some awesome systems like ROS (Robotic Operating System). Reply

Leave a Reply Your email address will not be published. Required fields are marked *