As the internet has evolved, and connectivity along with it, visuals have also increasingly become a key element that stands out, grabbing user attention in ever-busy social Feeds. This somewhat explains why memes are so popular right now.

It all started with once static images, then moved to GIFs, and now video has become the most engaging type of content. However, in essence, what you really need are engaging and interesting visuals that are eye-catching enough to stop people mid-scroll, which, for the most part, is far more effective than trying to catch them with blockbuster headlines and witty one-liners.

From Flat to Dynamic

This is what makes this news a bit interesting – Google had just outlined its latest 3D image creation process, called ‘LOLNeRF’, will be able to accurately estimate 3D structures from single 2D images. Check out this cool little sneak peek. Based on these examples, the LOLNeRF process takes your regular, 2D images and essentially ‘extrapolates’ them into 3D displays.

To note, Facebook has been offering a similar version for some time, but this new process by Google is far more advanced, enabling more depth and interactivity, minus the need to understand and capture full 3D models. As per Google:

“In “LOLNeRF: Learn from One Look”, we propose a framework that learns to model 3D structure and appearance from collections of single-view images. LOLNeRF learns the typical 3D structure of a class of objects, such as cars, human faces or cats, but only from single views of any one object, never the same object twice.”

The process can simulate color and density for each point in 3D space, by using ‘visual landmarks’ in the image, based on machine learning – it essentially replicates what the system knows from familiar images. From this, the process is then able to render more accurate, multi-dimensional visuals from a single, static source. This could have a huge range of applications, from AR art to expanded object creation in VR, as well as various Metaverse integrations.

True enough, if this process can accurately create 3D depictions of a wide assortment of 2D images, then that could greatly accelerate the development of 3D objects, helping to also quicken the development of the Metaverse and its advanced world-building capacities.

The Wrap

For the Metaverse to accurately create custom digital spaces, it’ll need to reference much of its assets from real-world objects, which act as source material that fuels this new creative approach. Imagine just feeding a catalog of web images into a system, then have it spit out fully 3D equivalents for use in promotions and such. If Google can translate the LOLNeRF process into more practical and accessible usage options, then that might just greatly propel its own AR and VR ambitions.

Subscribe to our ‘Bottoms Up!’ Newsletter. Get the latest social media blogs about news, updates, trends, and effective social media strategies to take your business to the highest level from Tristan Ahumada and Jeff Pfitzer.

Sources 

https://bit.ly/3Lg8gbo