The latest version of the AI-based framework by Nvidia generates impressive results.
The company has recently presented its latest experiment in machine learning for image creation called StyleGAN2, originally revealed at CVPR 2020. The new version based on the original StyleGAN build promises to generate a seemingly infinite number of portraits in an infinite variety of painting styles.
The video shows how you can create new portraits in no time by controlling the content, identity, expression, and pose of the subject. Users are also allowed to tweak the artistic style, color scheme, and appearance of brush strokes.
“Overall, our improved model redefines the state of the art in unconditional image modeling, both in terms of existing distribution quality metrics as well as perceived image quality,” paper Analyzing and Improving the Image Quality of StyleGAN states.
You can learn more here.