Google AI Tool Generates Imagery From Users’ MS Paint-Style Sketches
| Last updated
Bringing your drawings to life is difficult, particularly if you’re not skilled at illustration. Fortunately, Google has developed a tool that will breathe life into MS Paint-style images.
A tool called Chimera Painter allows users to have their sketches transformed into detailed images through machine learning.
Essentially the software adds texture and colour to the outlines that users sketch and some of the results that have been demonstrated, alongside the announcement of the software, have been amazing.
The team at Google explained exactly what the technology does in their announcement:
Today, we present Chimera Painter, a trained machine learning (ML) model that automatically creates a fully fleshed out rendering from a user-supplied creature outline. Employed as a demo application, Chimera Painter adds features and textures to a creature outline segmented with body part labels, such as “wings” or “claws”, when the user clicks the “transform” button.
The intention behind the technology was to make a ‘paintbrush that acted less like a tool and more like an assistant’. It seems that the team has achieved this, and to test the software they have applied the work to fantasy-based creatures to create a unique style for a video game.
To create this software, the team experimented with generative adversarial networks (GAN) which were used with artist feedback to make the images. The result was successful, but the team needed to adapt to the ‘potential for loss of anatomical and spatial coherence’ when using GAN and began incorporating other techniques to enable the software to add detail. Part of this process included using the Unreal Engine to generate 3D models renderings.
The card game that has been created showcases how Chimera Paint can speed up drawing processes while sticking to an artistic vision. Going forward, it is hoped that the tool can achieve a wider audience on the back of its demo.
The demo is now available, and Google explained what can be achieved with it:
The trained GAN is now available in the Chimera Painter demo, allowing artists to work iteratively with the model, rather than drawing dozens of similar creatures from scratch. An artist can select a starting point and then adjust the shape, type, or placement of creature parts, enabling rapid exploration and for the creation of a large volume of images. The demo also allows for uploading a creature outline created in an external program, like Photoshop.
This looks like a great way for artists to save time and for drawers with less skill to see their vision come to life.
If you have a story you want to tell, send it to UNILAD via [email protected]