Not a great artist? Don’t worry! This new Artist Bot under development by Microsoft researchers can transform your imagination into a reality using artificial intelligence.
Microsoft says this artistic bot or simply a drawing bot can create different images from “ordinary pastoral scenes, such as grazing livestock, to the absurd, such as a floating double-decker bus”.
This bot uses text description to generate images and according to Microsoft, the technology used can boost the quality of images three times more when compared to the state-of-art technique used previously for text-to-image generation.
“If you go to Bing and you search for a bird, you get a bird picture. But here, the pictures are created by the computer, pixel by pixel, from scratch,” explained Xiaodong He, a principal researcher and research manager in the Deep Learning Technology Center at Microsoft’s research lab in Redmond, Washington, in Microsoft’s announcement. “These birds may not exist in the real world — they are just an aspect of our computer’s imagination of birds.”
To match the words to images, the bot was trained on datasets of paired images and captions, says Microsoft. Also, it adds details that are not present in the description based on the imagination suggesting that this artificial intelligence has an artificial imagination.
The technology used in the Microsoft’s drawing bot is called Generative Adversarial Network, or GAN. It consists of two machine learning models – one to generate images from text description and another that uses text descriptions to judge the authenticity of generated images. The former tries to get fake pictures past the latter, while the latter never wants to be fooled. They together try to produce better high-quality images.
This technology can be used by painters and interior designers for sketch assistance. It could even be used to generate animated films based on screenplays according to He.
For further details, check the research paper.