Creating the high end fidelity of modern VR, AR or other PBR experiences takes hours and hours of your time with complex lighting gear and all sorts of other equipment. We didn't have time for all that and neither do you!
Our system uses AI trained on millions of images to create a network so smart it can accurately infer what you need directly from camera photos. Just upload your source photographs and download normals, roughness and de-lit texture maps (and more to come!).
Deep Physically Based Rendering uses AI to remove shadows and generate your image maps through deep learning. The DeepPBR Engine has learnt from other images and other textures, so that it can create normal maps, roughness maps and diffuse (delit) images from a simple photo, taken in normal light.
Once it has generated normals, roughness and diffuse images from the source image, it's a simple matter of rendering this in your favorite rendering pipeline. We demonstrate here with WebGL in your browser.
Below are some photos taken in normal light with a regular camera.
(Click on any image)
Models generated directly from the source images using our (early alpha) Deep PBR pipeline.
(Click and drag to rotate)
Chief Visionary Officer