Artificial intelligence learned to make full-fledged 3D animation from photography

Information has appeared online about a recent development by a group of researchers from Google. Thanks to them, artificial

intelligence has learned to turn a photograph into a full-fledged 3D animation. However, it is worth noting that for a decent result it is necessary to use high-quality photographs.

As part of a new project, researcherspresented detailed 3D scenes of famous landmarks around the world. These included the Brandenburg Gate in Berlin, the Sacre Coeur Basilica in Paris and the Trevi Fountain in Rome. Interestingly, thanks to the use of machine learning, it has become possible not only to create beautiful 3D animation based on a photograph – specialists were able to recreate full-fledged three-dimensional scenes with the ability to move a virtual camera, and the external scene, at the same time, can change under the influence of lighting effects.

At the same time, the developers clarify that at this timeAt this stage, their development is extremely capricious: “Two photographers can stand in the same place and take a picture of the same attraction, but the algorithm will process one frame correctly, and spoil the second. We need to improve the AI ​​so that it works correctly,” – said one of the study authors.