EA Could Create In-Game Character Animations Using AI To Replace Motion Capture

Using AI may aid devs to replace the use of motion capture technology in games to some extent.

Story Highlights

  • EA has published a new patent that seeks to generate in-game character animations automatically on the spot using AI. It seeks to replace motion capture where it is not feasible.
  • The patent will let AI generate animations in-game environments where motion capture is difficult or impossible to use. The system can create the following frames of animations.
  • The system can be used to replace motion capture, where it can harm the actor. It can also reduce the amount of manual animation created by devs to make it more automatic.
  • The patent may allow EA to generate more dynamic animations for games using AI to ensure every environment or variable can be accounted for in the game character.

EA has produced legal documents in the past for automating a variety of things that concern game development to help devs with tedious tasks. A recent patent by the giant conglomerate suggests that it may also automate in-game animations using AI to replace motion capture, at least in cases where it is not suitable.

The patent dubbed “DYNAMIC LOCOMOTION ADAPTATION IN RUNTIME GENERATED ENVIRONMENTS talks about utilizing AI models to predict the frames of animations for in-game characters during gameplay in environments where it is nearly impossible for motion capture to work. It argues that the current system requires a lot of motion capture data that takes up a lot of space, and it is not viable in all cases to be used for animation.

It may be difficult to predict all the possible interactions that may occur in the video game making it difficult to predetermine all the motion capture data that is required during design and development of the video game,” argues EA.

The image shows automatica frame generation by the dynamic animation system using AI.
The image shows automatic frame generation by the dynamic animation system using AI.

This dynamic animation generation system can use motion capture data to train and predict the subsequent frames to produce in-game character animations using AI. It is quite beneficial to use deep learning technology because motion capture data is not reliable in certain game environments and for specific in-game characters. For instance, a cartoony character not resembling a human cannot effectively use motion data to be animated.

The in-game environments include places that are not easily compatible with motion capture in games. EA has noted some instances like, “underwater environments, non-Earth based gravity environments, non-Earth based atmosphere environments, non-human life supporting temperature environments, etc.

The system may use the automated analyses of the motion capture information to extend the generation of realistic motion to new environments that are difficult if not impossible […] to capture using motion capture technology.”

The image shows a detailed block diagram of the dynamic animation system.
The image shows a detailed block diagram of the dynamic animation system.

It is also worth noting that motion capture studios can only create a limited set of animations for a game character. For instance, the patent gives the example of the motion capture recorder having to run and slip on ice to generate realistic animations for the character, which is harmful to the person and cost extensive. It would also not be possible for a character to do actions of the same quality that are not recorded by motion capture.

These extra motions that were not recorded could be created using manual methods by the game designer, but the new patent by EA seeks to solve that dilemma by using AI to produce character animations automatically during the gameplay.

Advantageously, the environment properties provided as input to the prediction model may differ from the environment properties included in the training data. Thus, animation can be generated that accounts for new environments for which motion capture data is not available.”

The image shows a detailed flowchart diagram to explain the working of the dynamic animation system.
The image shows a detailed flowchart diagram to explain the working of the dynamic animation system.

EA argues that using AI will also make transitioning between various in-game biomes or their blends faster for in-game characters. The animations could be created by AI on the spot to account for many variables together. For instance, the hot or cold, elevation or gravity levels, and much more could impact how the animations are created in certain settings using AI instead of manually accounting for all instances using motion capture.

Electronic Arts’ new patent could help game devs to utilize AI alongside motion capture to create animations more easily, as it would also reduce the tediousness and the storage space required to record all animations manually. Moreover, it may help EA to feature more dynamic animations in its games, which is not possible by current methods.

EA has published some eccentric patents in the past, including a videographer role that would let users record the gameplay in the middle of online matches, a system to populate in-game maps with structures using the wonders of AI, and many more.  

Similar Reads: MachineGames Working On An Unannounced Title Other Than Indiana Jones.

Did you find this article helpful?

Thanks! Do share your feedback with us. ⚡

How can we make this post better? Your help would be appreciated. ✍

Get up-to-speed gaming updates delivered right to your inbox.

We don’t spam! Read more in our privacy policy.

Source
Patentscope

Shameer Sarfaraz is a Senior News Writer on eXputer who loves to keep up with the gaming and entertainment industries devoutly. He has a Bachelor's Degree in Computer Science and several years of experience reporting on games. Besides his passion for breaking news stories, Shahmeer loves spending his leisure time farming away in Stardew Valley. VGC, IGN, GameSpot, Game Rant, TheGamer, GamingBolt, The Verge, NME, Metro, Dot Esports, GameByte, Kotaku Australia, PC Gamer, and more have cited his articles.

Experience: 4+ Years || Education: Bachelor in Computer Science.

Related Articles