Relightable scene reconstructions using radiance guided material extraction

Abstract

Techniques for relightable scene reconstructions using radiance guided material extraction are described to accurately render 3D scenes under different lighting conditions and perspectives than original source images from which the scenes are constructed. In an example, a processing device is operable to receive a plurality of digital images that depict a scene from multiple perspectives, determine a view-independent radiance of the scene based on the plurality of digital images, and determine a view-dependent radiance of the scene based on the plurality of digital images. The processing device is further operable to determine a set of lighting conditions associated with an input perspective, generate a synthesized image having a reconstruction of the scene based on the set of lighting conditions using the view-independent radiance and the view-dependent radiance, and output the synthesized image.