Nvidia’s secret weapon to win the race for the metaverse

While the metaverse is now a priority for many technology giants, Nvidia is taking advantage of a new keynote dedicated to graphic technologies to flex its muscles against the competition. There’s a reason: Nvidia has actually started the race for the metaverse by presenting the latest technologies to make this digital universe a reality. The company – like the entire sector – nonetheless faces a formidable problem: how to transform the 2D world into 3D without going through complicated, forced and – most of all – slow processes.

The American giant may have seen the parade to promote this thorny operation. unsaon? Thanks to a new method of reverse rendering-the process of reconstructing 3D scenes from several 2D images. With its new approach, Nvidia uses artificial intelligence to estimate the behavior of light as it would in the real world. Even better: in the manner employed by the technology giant’s teams, the whole process now happens almost immediately.

According to Nvidia, “this technology can be used to teach robots and self-driving vehicles to understand the size and shape of real-world objects. […] or in the field of architecture and entertainment to easily create digital representations of real -world environments that can be modified and expanded by creators. Something that will be of interest to metaverse players, who are now working hard to establish themselves in this market in the future.

Revolutionary technology

The latest technology designed by Nvidia could be a breakthrough in the metaverse needed to provide itself with the means to achieve its ambitions. To date, using traditional methods to create a 3D scene can take several hours depending on the complexity and resolution of the visualization. However, artificial intelligence has greatly shortened the process by taking advantage of a popular new technology called Neural Radiance Fields (NeRF).

These NeRFs are based on neural networks to represent and render realistic 3D scenes from a collection of 2D images. A NeRF effectively trains a small neural network to fill in the blanks by predicting the color of light shining in any direction from any point in 3D space. The first NeRF models already provided quality translations in just a few minutes, but training them took hours.

This is exactly where Nvidia’s research teams come into play. They created an instant NeRF, combining fast neural network training and fast translation. According to Nvidia’s management, this is the fastest NeRF technique to date, with speeds in excess of 1,000 times in some cases. To improve Instant NeRF, Nvidia specifically relied on a new input encoding method called multi-resolution hash grid encoding. This method is optimized to work effectively with Nvidia GPUs.

“Instant NeRF can be just as important in 3D as digital camera and JPEG compression than 2D photography, dramatically increasing the speed, ease and reach of 3D capture and sharing,” said David Luebke, vice president of graphics research at Nvidia. It remains to be seen what Nvidia intends to do with this new technology that could allow it to consolidate its weight in the new metaverse market.

Leave a Comment