GauGan2 Showcases New AI Art Demo

GauGAN2, the newest version of the AI painting demo of NVIDIA Research, enables users to produce photorealistic images by combining segmentation mapping, inpainting and now, text-to-image generation.

The demo is one of the first to combine multiple modalities — text, semantic segmentation, sketch and style — within a single GAN framework. This makes it faster and easier to turn an artist’s vision into a high-quality AI-generated image.

Through the use of GauGAN2’s text-to-image feature, users can now produce the images in their minds. To do that, a user only needs to type phrases like “sunset at a beach” and the AI generates the scene in realtime. This is achieved through the deep learning model of GauGAN that is based on the generative adversarial networks, which also allows the instant modification of the AI rendered picture by just including additional adjectives like “sunset at a rocky beach”.

GauGAN2 also allows users to tweak the scene to their liking. With just a press of a button, a segmentation map can be generated to show the location of objects in the scene. This allows users to draw on, or tweak the scene with rough sketches using labels such as sky, tree, rock, or river, which will allow the smart paintbrush to incorporate them into the image.

GauGAN2 is a powerful tool to create photorealistic art with a mix of words and drawings. To experience the wonder of the AI painting demo, users just need to click on this link – NVIDIA AI Demos. Aside from GauGan2, the site also offers the latest AI demos from NVIDIA Research.

You can learn more about NVIDIA Research and other NVIDIA AI products by visiting the official NVIDIA website. You may also follow their official Facebook page for more News about the brand.

You can also stay updated to news about the brand by following our page and by regularly checking on our website.