The world is full of a number of ‘what ifs’, making everything from new factory layouts to autonomous vehicle driving to coastal tsunami defence difficult at best. However, what if these ‘what ifs’ could be solved in a parallel virtual world before work started in our own?
What, too, if this parallel world worked fast enough to give near-instant predictions? This would mean that everything from Atlantic hurricanes to Californian wildfires could be predicted more accurately, giving those on the ground real-time advice.
If all is to be believed, this world will soon be with us though NVIDIA’s Omniverse Replicator software, which Time is calling one of the best inventions of 2021.
In 1999 NVIDIA sparked the growth of the PC gaming market by inventing GPU which redefined modern computer graphics, and revolutionized parallel computing. More recently, GPU deep learning ignited modern AI – the next era of computing – with the GPU acting as the brain of computers, robots, and self-driving cars that can perceive and understand the world.
According to Jesse Will of Time, the Omniverse Replicator project is making it easier to create ultra-realistic virtual spaces for those real-world purposes. The platform combines the real-time ray-tracing technology of the brand’s latest graphics processing units with an array of open-source tools for collaborating live in photorealistic 3-D worlds. Architects like Foster + Partners are using the tech to visualize building details earlier in the design process, and BMW has built a “digital twin” of one of its factories, enabling the company to test more efficient assembly-line possibilities.
“Omniverse Replicator allows us to create diverse, massive, accurate datasets to build high-quality, high-performing and safe datasets, which is essential for AI,” said Rev Lebaredian, vice president of simulation technology and Omniverse engineering at NVIDIA. “While we have built two domain-specific data-generation engines ourselves, we can imagine many companies building their own with Omniverse Replicator.”
Omniverse Replicator augments costly, laborious human-labelled real-world data, which can be error prone and incomplete, with the ability to create large and diverse physically accurate data tailored to the needs of AV and robotics developers. It also enables the generation of ground truth data that is difficult or impossible for humans to label, such as velocity, depth, occluded objects, adverse weather conditions or tracking the movement of objects across sensors.
Already an invaluable data-generation engine for NVIDIA’s DRIVE autonomous vehicle and Isaac robotics teams, Omniverse Replicator will be made available next year to developers to build domain-specific data-generation engines.
For more on NVIDIA’s Omniverse Replicator, listen to NVIDIA’S CEO Jensen Huang explain its potential during the GTC November 2021 Keynote talk available on YouTube.
