Current Issue
This Month's Print Issue

Follow Fast Company

We’ll come to you.

1 minute read


Microsoft Research Debuts Autocomplete For Animation, And It’s Incredible

Draw something once, and software can predict, guide, and adjust what you want to do next.

  • 01 /05

    Microsoft Research Asia has a new animation technology.

  • 02 /05

    It's autocomplete for drawing, essentially.

  • 03 /05

    You draw something once, and it predicts your next frame.

  • 04 /05

    As you make adjustments, its suggestions will adjust, too.

  • 05 /05

    If built into a product, the technology could make animation via simple tablet interfaces extremely simple.

Microsoft Research, working with the University of Hong Kong and the University of Tokyo, has a remarkable new technology that it calls "Autocomplete hand drawn animations."

Unveiled at the Siggraph Asia conference, it doesn’t appear that a formal white paper has been released, but there is this illustrative video of the tech. You can watch as someone draws a fish once, then, upon drawing a single line for the next frame, the software suggests a skeleton to trace. It’s responsive, too, shaping its wireframes around your sketches in real time. And it also features a few other neat tricks, like if you begin filling in a pattern, it can sense that and suggest more of the pattern—or if you add color or detail to the first frame of your animation, it can duplicate it across the entire work.

Microsoft isn’t the only company playing in this space. Autodesk’s Draco allows you to animate still images with quick motion elements. Adobe is able to create 3-D models from 2-D images. And Disney has been using semi-automated technology called Meander to fill in frames of animation since their Oscar-winning short, Paperman. (And even in the earlier days of Disney, low level animators were tasked with this job of drawing fill frames, while the elite animators would draw establishing start and end points.)

In this sense, Microsoft Research is just further-automating animation tradition, turning human labor into computer labor. But what’s most incredible about the demo isn’t just the software’s ability to predict or suggest sketches, but to reshape its suggestions in real time as you draw—and to do so without complicated interface modalities like anchor points or keyframe markers. It’s just the sort of quick, menu-bar-less interface that feels perfect for tablets—assuming Microsoft ever brings the technology to market.

[via Prosthetic Knowledge]

Related: The Evolution Of Microsoft In 3 Minutes

The Fast Company Innovation Festival