Shulou(Shulou.com)11/24 Report--

Thanks to CTOnews.com netizen Sancu for the clue delivery! CTOnews.com June 2, Nvidia announced this week the launch of an AI model called Neuralangelo, which converts video clips into 3D models with rich levels of detail and textures.

▲ image source Nvidia CTOnews.com query learned that there have been several AI that generate 3D models based on video content, but these AI are often unable to deal with low-contrast video. Nvidia's Neuralangelo model uses real-time neurographics (instant neural graphics primitives) and its own Instant NeRF technology, so even for low-contrast videos, subtle details can be captured and converted into more commercially valuable 3D models.

▲ source Yingweida Nvidia stressed that the uniqueness of Neuralangelo mainly lies in its "accuracy" and "universality":

"accuracy" means that the model can accurately present the texture of complex materials, such as roof tiles, glass panes and smooth marble.

"universality" means that creators can simply use the film material shot by smartphones to create realistic objects and integrate them into their projects.

The ▲ image source Nvidia Neuralangelo model uses videos of objects taken from different angles as input material to generate the models needed by creators. Once Neuralangelo determines the motion pattern of the camera in the video, it generates a 3D outline of the model and begins to carve details on it.

▲ Tu Yuan Nvidia expects to unveil the Neuralangelo model at the computer Vision and style recognition (Computer Vision and Pattern Recognition,CVPR) conference on June 18-22. In the internal demonstration, Nvidia showed 3D restoration models of everyday objects such as Michelangelo's famous sculpture David, Nvidia's office in the Bay area, and flat trailers reconstructed from the Neuralangelo model.

▲ Source in addition to Neuralangelo, Nvidia will also show a model called DiffCollage at the conference. DiffCollage is an AI model that can generate video through text prompts. It can generate 360-degree panoramic images and cyclic dynamic images, and then expand these images to specific videos.

Ming-Yu Liu, senior director of research at Nvidia and author of the paper, imagined the significance of the research:

The 3D reconstruction capabilities provided by Neuralangelo will bring great benefits to creators and help them rebuild the real world in the digital world. This tool will eventually enable developers to import fine objects-whether small statues or large buildings-into video games or industrial digital twins.

CTOnews.com 's friends can find the specific content of the paper here.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.