Abstract: Rendering an avatar in a selected environment may include determining as inputs into an inferred shading network, an expression geometry to be represented by an avatar, head pose, and camera angle, along with a lighting representation for the selected environment. The inferred shading network may then generate a texture of a face to be utilized in rendering the avatar. The lighting representation may be obtained as lighting latent variables which are obtained from an environment autoencoder trained on environment images with various lighting conditions.

Abstract: Various implementations disclosed herein include devices, systems, and methods that generate exposed panoramic image data of an environment. For example, an example process may include obtaining panoramic image data including pixel values corresponding to light received by one or more image capture devices from multiple angles in a physical environment, generating blurred panoramic image data by adjusting the individual pixel values based on values of a subset of other values of the pixel values, selecting an area of the blurred panoramic image data, and determining an exposure for the panoramic image data based on the area of the blurred panoramic image data.

Abstract: Rendering an avatar in a selected environment may include determining as inputs into an inferred shading network, an expression geometry to be represented by an avatar, head pose, and camera angle, along with a lighting representation for the selected environment. The inferred shading network may then generate a texture of a face to be utilized in rendering the avatar. The lighting representation may be obtained as lighting latent variables which are obtained from an environment autoencoder trained on environment images with various lighting conditions.

Abstract: Image-based lighting for use in rendering of digital objects is derived from image data collected from a real motion-picture set. The image data includes sets of correlated images at different exposure levels, each corresponding to a selected location. The image data is processed to define an extended dynamic range panoramic image for each selected set location. The image data is color and intensity corrected based on a scaled reference object. At least key lights are modeled for the set by processing a plurality of such panoramic images. Other lighting may be defined by interpolating from the panoramic images to mathematically define a location-specific set of fill lights. In the alternative, other set lighting may be modeled by projecting panoramic image data onto a dynamically sub-dividable light primitive derived from the set geometry. Any desired object may then rendered using the defined lights, and inserted into a corrected real image of the set at the desired location.

Abstract: Image-based lighting for use in rendering of digital objects is derived from image data collected from a real motion-picture set. The image data includes sets of correlated images at different exposure levels, each corresponding to a selected location. The image data is processed to define an extended dynamic range panoramic image for each selected set location. The image data is color and intensity corrected based on a scaled reference object. At least key lights are modeled for the set by processing a plurality of such panoramic images. Other lighting may be defined by interpolating from the panoramic images to mathematically define a location-specific set of fill lights. In the alternative, other set lighting may be modeled by projecting panoramic image data onto a dynamically sub-dividable light primitive derived from the set geometry. Any desired object may then rendered using the defined lights, and inserted into a corrected real image of the set at the desired location.