NVIDIA Changes the AI Game with Words That Become Images in Real Time

In the world of artificial intelligence, NVIDIA has been a constant strong player shelling out new tech and developing complex algorithms that allow people to create amazing things with ease. Previously the tech company has unveiled an AI that can match voice lines with facial expressions of 3D animated faces. Today, NVIDIA's AI innovation uses words to generate photorealistic images in real-time.

What is AI?

(Photo: Martin Lopez from Pexels)

According to IBM, Artificial Intelligence, or AI, uses machines and computers to mimic the unique problem-solving and decision-making skills of humans. Over the past decades, numerous definitions for AI have sprouted; however, according to a paper published in 2004 by John McCarthy, AI is the science of making intelligent machines unbound by biologically observable methods.

There are two types of AI--strong and weak AI. Weak AI is trained to perform specific tasks consisting of most AI that surrounds us today. Apple's Siri, IBM's Watson, Amazon's Alexa, and even autonomous vehicles are examples of Weak AI.

On the other hand, Strong AI is made of Artificial General Intelligence and Artificial Super Intelligence. It is the embodiment of AI's theoretical form where machines would have the equal intelligence of humans, be self-aware, and have the ability to learn, solve problems, and plan for the future.

ALSO READ: Mystery on Perovskite Solar Cell's Tolerance to Defects Uncovered For the First Time

NVIDIA's Revolutionary AI: GuaGAN2

A few years back, a simpler form of NVIDIA's AI-- GuaGAN first made headlines as it demonstrated how it could turn basic doodles into images with only a few clicks of a button. It attracted numerous tech-heads and gamers enticed by its unique appeal and then advanced AI.

Today GuaGAn2 is available and has advanced so much that it no longer requires sketches to create highly detailed photorealistic landscape images. Instead, NVIDIA's AI and deep learning model turns text into surprisingly complex photorealistic images. The AI was trained on over 10 million landscape images. NVIDIA demonstrates how images generated by the AI change as more text is added, implying that it can adapt and make changes in real-time.

NVIDIA's AI GuaGAN 2 works using fundamentals seen in GuaGAN that creates images based on shapes, combined with the advancements of text-to-image generation. This means that a mix of drawings and words can be used in GuaGAN 2 to create stunning photorealistic images. Users can now start their AI journey with a general statement and add words or shapes as they go on to get more detailed landscape images from NVIDIA"s AI.

To enjoy the recent AI innovations of NVIDIA, all you need is an NVIDIA RTX GPU. It is available for free download via NVIDIA Canvas. In terms of game implications, the technology is sure to cut downtime for game developers since users can find reference images online and do the bare minimum to replicate them into words or shape. NVIDIA'S AI Canvas will do the rest in real-time.