Abstract: A learning agent is disclosed that receives data in sequence from one or more sequential data sources; generates a model modelling sequences of data and actions; and selects an action maximizing the expected future value of a reward function, wherein the reward function depends at least partly on at least one of: a measure of the change in complexity of the model, or a measure of the complexity of the change in the model. The measure of the change in complexity of the model may be based on, for example, the change in description length of the first part of a two-part code describing one or more sequences of received data and actions, the change in description length of a statistical distribution modelling, the description length of the change in the first part of the two-part code, or the description length of the change in the statistical distribution modelling.

Abstract: A lighting stage includes a plurality of lights that project alternating spherical color gradient illumination patterns onto an object or human performer at a predetermined frequency. The lighting stage also includes a plurality of cameras that capture images of an object or human performer corresponding to the alternating spherical color gradient illumination patterns. The lighting stage also includes a plurality of depth sensors that capture depth maps of the object or human performer at the predetermined frequency. The lighting stage also includes (or is associated with) one or more processors that implement a machine learning algorithm to produce a three-dimensional (3D) model of the object or human performer. The 3D model includes relighting parameters used to relight the 3D model under different lighting conditions.

Abstract: An example method, apparatus, and computer-readable storage medium are provided to predict high-dynamic range (HDR) lighting from low-dynamic range (LDR) background images. In an example implementation, a method may include receiving low-dynamic range (LDR) background images of scenes, each LDR background image captured with appearance of one or more reference objects with different reflectance properties; and training a lighting estimation model based at least on the received LDR background images to predict high-dynamic range (HDR) lighting based at least on the trained model. In another example implementation, a method may include capturing a low-dynamic range (LDR) background image of a scene from an LDR video captured by a camera of the electronic computing device; predicting high-dynamic range (HDR) lighting for the image, the predicting, using a trained model, based at least on the LDR background image; and rendering a virtual object based at least on the predicted HDR lighting.

Abstract: Methods, systems, and media for relighting images using predicted deep reflectance fields are provided.

Abstract: Methods, systems, and media for relighting images using predicted deep reflectance fields are provided.

Abstract: Various systems and methods disclosed herein are directed to rendering point-based graphics on a computing device with the spaces between points filled in with color to produce the appearance of surfaces without gaps or holes. According to one method, one or more rasterization passes are performed on an image space. One or more filling passes are performed on the pixels in the image space in which the spaces between the pixels are filled with color to form a contiguous surface in a new image plane. One or more blending passes are performed on the image space after the filling passes, in which wherein the color of a group of pixels is a blended together. A new image space is rendered from the image space in the image buffer.

Abstract: A learning agent is disclosed that receives data in sequence from one or more sequential data sources; generates a model modelling sequences of data and actions; and selects an action maximizing the expected future value of a reward function, wherein the reward function depends at least partly on at least one of: a measure of the change in complexity of the model, or a measure of the complexity of the change in the model. The measure of the change in complexity of the model may be based on, for example, the change in description length of the first part of a two-part code describing one or more sequences of received data and actions, the change in description length of a statistical distribution modelling, the description length of the change in the first part of the two-part code, or the description length of the change in the statistical distribution modelling.

Abstract: A system and methods for suggesting beneficial actions may include generating a first predictive model to estimate a user profile, collecting user data pertinent to a user's circumstance or action at a point in time, collecting a database associating user profiles with sequences of user circumstances and user actions over time for one or more users, and generating a second predictive model to suggest an action or circumstance or a sequence of actions and circumstances to be communicated to a user based upon the estimated user profile and user data and the database of sequences of user circumstances and actions. The system delivers the suggested action or circumstance to a device carried by or worn by a user in the form of advice or a story, using text, speech, images, or video.

Abstract: Various systems and methods disclosed herein are directed to rendering point-based graphics on a computing device with the spaces between points filled in with color to produce the appearance of surfaces without gaps or holes. According to one method, one or more rasterization passes are performed on an image space. One or more filling passes are performed on the pixels in the image space in which the spaces between the pixels are filled with color to form a contiguous surface in a new image plane. One or more blending passes are performed on the image space after the filling passes, in which wherein the color of a group of pixels is a blended together. A new image space is rendered from the image space in the image buffer.

Abstract: A multiview face capture system may acquire detailed facial geometry with high resolution diffuse and specular photometric information from multiple viewpoints. A lighting system may illuminate a face with polarized light from multiple directions. The light may be polarized substantially parallel to a reference axis during a parallel polarization mode of operation and substantially perpendicular to the reference axis during a perpendicular polarization mode of operation. Multiple cameras may each capture an image of the face along a materially different optical axis and have a linear polarizer configured to polarize light traveling along its optical axis in a direction that is substantially parallel to the reference axis. A controller may cause each of the cameras to capture an image of the face while the lighting system is in the parallel polarization mode of operation and again while the lighting system is in the perpendicular polarization mode of operation.

Abstract: A controllable lighting system may include a plurality of light source groups, a group controller for each light source group, a master controller, and a network communication system. Each group controller may be configured to control the light sources in its light source group based on a group control command. The master controller may be configured to receive a master control command relating to the light sources and to issue a group control command to each of the group controllers that collectively effectuate compliance with the master control command. The network communication system may be configured to communicate the group control commands from the master controller to the group controllers.

Abstract: A system for estimating the specular roughness of points on a surface of an object may include a lighting system, an image capture system and a computer processing system. The lighting system may be configured to illuminate the surface of the object at different times with different illumination patterns. Each illumination pattern may illuminate the surface from a plurality of different directions and form an intensity gradient having an order of no more than two. The image capture system may be configured to capture an image of the surface of the object when illuminated by each of the different illumination patterns at each of the different times. The computer processing system may be configured to compute the specular roughness of each point on the surface of the object based on the images captured by the image capture system.

Abstract: A multiview face capture system may acquire detailed facial geometry with high resolution diffuse and specular photometric information from multiple viewpoints. A lighting system may illuminate a face with polarized light from multiple directions. The light may be polarized substantially parallel to a reference axis during a parallel polarization mode of operation and substantially perpendicular to the reference axis during a perpendicular polarization mode of operation. Multiple cameras may each capture an image of the face along a materially different optical axis and have a linear polarizer configured to polarize light traveling along its optical axis in a direction that is substantially parallel to the reference axis. A controller may cause each of the cameras to capture an image of the face while the lighting system is in the parallel polarization mode of operation and again while the lighting system is in the perpendicular polarization mode of operation.

Abstract: A controllable lighting system may include a plurality of light source groups, a group controller for each light source group, a master controller, and a network communication system. Each group controller may be configured to control the light sources in its light source group based on a group control command. The master controller may be configured to receive a master control command relating to the light sources and to issue a group control command to each of the group controllers that collectively effectuate compliance with the master control command. The network communication system may be configured to communicate the group control commands from the master controller to the group controllers.

Abstract: A system for estimating the specular roughness of points on a surface of an object may include a lighting system, an image capture system and a computer processing system. The lighting system may be configured to illuminate the surface of the object at different times with different illumination patterns. Each illumination pattern may illuminate the surface from a plurality of different directions and form an intensity gradient having an order of no more than two. The image capture system may be configured to capture an image of the surface of the object when illuminated by each of the different illumination patterns at each of the different times. The computer processing system may be configured to compute the specular roughness of each point on the surface of the object based on the images captured by the image capture system.