Like Nvidia in one Blog entry He writes, the latest NeRF approach takes only a few seconds to train a model. After creating a 3D scene, only milliseconds will be needed to display the images from the scene. The technology relies on multiple images that capture the object in question from different positions. However, if the images contain several objects, they should not move too fast. Otherwise there is a risk that the result will be falsified.
The developers call their alternative “Instant NeRF” and confirm that it is currently the fastest method. Traditional methods take several hours to create a 3D scene. Instant NeRF is said to be able to complete tasks 1,000 times faster than other NeRF methods. The technology uses artificial intelligence to identify shadows within the image and create a model for display.
Can be used in many areas
Instant NeRF could be used in the future to improve self-driving vehicles. With this technology, systems can recognize and understand the sizes and shapes of objects in the real world. In addition, the process is likely to play a role in the leisure and architecture sector. Several photos will be enough to create a 3D model of the environment. This can save a lot of time for developers and architects.
“Travel maven. Beer expert. Subtly charming alcohol fan. Internet junkie. Avid bacon scholar.”