The world of technology is moving in two directions for all users. On the one hand, we have 3D technology, which is being strongly implemented in the world of modeling and printing. On the other hand, there is artificial intelligence, capable of doing all kinds of things with a couple of samples of all kinds. The last of both houses has broughto NVidia by showing an AI capable of creating a three-dimensional image with a few photos.
Goodbye to multicapture, with few photos you have one in 3D
To obtain a three-dimensional image, a machine compatible with this technology and another capable of reading it with these characteristics is needed. However, it is impossible to believe that a machine is capable of doing this job, especially since if you use it to take the same photo from several angles you will lose data. To further illustrate this, the famous Matrix sequence where Neo dodged bullets in slow motion was achieved with multiple cameras taking multiple photos from multiple angles.
But at NVidia they have come up with a very interesting solution to interpret several 2D images and generate one in 3D. As you can see in the video, the firm references four photos taken at the same time in the same place. However, artificial intelligence does the interpretation work to have in a file an image that can be seen in 3D.
In other words, through the images that you offer the computer, it makes an image taking the position of the objects as a reference and you can navigate through it fluently. In this case, the woman can be seen as if you were walking around her but without losing any details thanks to the work of artificial intelligence.
Instant NeRF, a rendering technology in seconds
The other good part of this artificial intelligence from Nvidia is the time it takes to take an image of this style. With a traditional computer, as we were saying, this task involves using a lot of resources and, of course, a lot of rendering time. But the system Instant NeRF it is much more powerful than that since it stands out because in a matter of seconds the AI has the file.
“In that sense, Instant NeRF could be as important to 3D as digital cameras and JPEG compression have been to 2D photography, vastly increasing the speed, ease and scope of 3D capture and sharing.” “, commented the company itself.