Lip syncing for computer animated characters has long been simplified. You draw a set of lip shapes for vowels and other sounds your character makes and let the computer interpolate how to go from one shape to the next. But with physical, real world puppets, all those movements have to be done manually, frame-by-frame. Or do they?
Stop motion animator and maker/hacker [James Wilkinson] is working on a project involving a real-world furry cat character called Billy Whiskers and decided that Billy’s lips would be moved one frame at a time using servo motors under computer control while [James] moves the rest of the body manually.
He toyed around with a number of approaches for making the lip mechanism before coming up with one that worked the way he wanted. The lips are shaped using guitar wire soldered to other wires going to servos further back in the head. Altogether there are four servos for the lips and one more for the jaw. There isn’t much sideways movement but it does enough and lets the brain fill in the rest.
On the software side, he borrows heavily from the tools used for lip syncing computer-drawn characters. He created virtual versions of the five servo motors in Adobe Animate and manipulates them to define the different lip shapes. Animate then does the interpolation between the different shapes, producing the servo positions needed for each frame. He uses an AS3 script to send those positions off to an Arduino. An Arduino sketch then uses the Firmata library to receive the positions and move the servos. The result is entirely convincing as you can see in the trailer below. We’ve also included a video which summarizes the iterations he went through to get to the finished Billy Whiskers or just check out his detailed website.
[Jame’s] work shows that there many ways to do stop motion animation, perhaps a part of what makes it so much fun. One of those ways is to 3D print a separate object for each character shape. Another is to make paper cutouts and move them around, which is what [Terry Gilliam] did for the Monty Python movies. And then there’s what many of us did when we first got our hands on a camera, move random objects around on our parent’s kitchen table and shoot them one frame at a time.
Citroën vient en aide aux personnes qui sont malades en voiture et lancent les lunettes Seetroën qui, en plus d'offrir un magnifique jeu de mots, permettent d'éliminer le mal des transports.
But that’s obviously just the beginning. As the DeepLens team stressed during our workshop, even developers who have never worked with machine learning can take the existing templates and easily extend them. In part, that’s due to the fact that a DeepLens project consists of two parts: the model and a Lambda function that runs instances of the model and lets you perform actions based on the model’s output. And with SageMaker, AWS now offers a tool that also makes it easy to build models without having to manage the underlying infrastructure.
