{"id":339769,"date":"2017-05-02T02:14:52","date_gmt":"2017-05-01T18:14:52","guid":{"rendered":"http:\/\/techcrunch.com\/?p=1483745"},"modified":"2017-05-02T02:14:52","modified_gmt":"2017-05-01T18:14:52","slug":"this-neural-network-could-make-animations-in-games-a-little-less-awkward","status":"publish","type":"post","link":"https:\/\/people.utm.my\/asmawisham\/this-neural-network-could-make-animations-in-games-a-little-less-awkward\/","title":{"rendered":"This neural network could make animations in games a little less awkward"},"content":{"rendered":"

<\/p>\n

The graphical fidelity of games these days is truly astounding, but one thing their creators struggle to portray is the variety and fluidity of human motion. An animation system powered by a neural network<\/a> drawing from real motion-captured data may help make our avatars walk, run, and jump a little more naturally.<\/p>\n

Of course, if you\u2019ve played any modern game, you may find that many do already \u2014 but this is painstaking work accomplished by animators working from libraries of motions, linking together all kinds of contingencies \u2014 what if she raises her bow as she\u2019s going up stairs and<\/em> crouching? What if she gets hit while trying to balance on a narrow beam? The possibilities are quite endless.<\/p>\n

Researchers from the University of Edinburgh and Method Studios put together a machine learning system that feeds on motion capture clips showing various kinds of movement. Then, when given an input a user saying \u201cgo this way\u201d and taking into account the terrain, it outputs the animation that best fits both \u2014 for example, going from a jog to hopping over a small obstacle.<\/p>\n

\"\"<\/a>No custom animation has to be made transitioning from a jog to a hop; the algorithm determines that, producing smooth movements and no jarring switches from one animation type to another. Although plenty of game engines do improvise a little for things like foot placement and blend animations, this is a new approach that could prove more robust.<\/p>\n

Machine learning has been brought to this space before, but as the team shows in the video, the systems produced were pretty rudimentary, showing the wrong type of movement or skipping animations because they weren\u2019t sure which to use \u2014 or deciding too strongly for an animation state and producing jittery motion.<\/p>\n