Abstract:
A camera system captures images from a set of cameras to generate binocular panoramic views of an environment. The cameras are oriented in the camera system to maximize the minimum number of cameras viewing a set of randomized test points. To calibrate the system, matching features between images are identified and used to estimate three-dimensional points external to the camera system. Calibration parameters are modified to improve the three-dimensional point estimates. When images are captured, a pipeline generates a depth map for each camera using reprojected views from adjacent cameras and an image pyramid that includes individual pixel depth refinement and filtering between levels of the pyramid. The images may be used generate views of the environment from different perspectives (relative to the image capture location) by generating depth surfaces corresponding to the depth maps and blending the depth surfaces.
Abstract:
A camera system captures images from a set of cameras to generate binocular panoramic views of an environment. The cameras are oriented in the camera system to maximize the minimum number of cameras viewing a set of randomized test points. To calibrate the system, matching features between images are identified and used to estimate three-dimensional points external to the camera system. Calibration parameters are modified to improve the three-dimensional point estimates. When images are captured, a pipeline generates a depth map for each camera using reprojected views from adjacent cameras and an image pyramid that includes individual pixel depth refinement and filtering between levels of the pyramid. The images may be used generate views of the environment from different perspectives (relative to the image capture location) by generating depth surfaces corresponding to the depth maps and blending the depth surfaces.
Abstract:
An online system identifies a conversion of an advertisement when a client device associated with an online system user is within a threshold distance of a physical location associated with the advertisement. The client device sends obfuscated information identifying the user associated with the client device and the location of the client device to an independent third party when the client device is within a threshold distance of the physical location associated with the advertisement. Using information from the client device, the third party requests information from the online system identifying one or more groups including the user. Based on information identifying groups including various online system users, the third party generates information identifying conversions associated with different groups. The online system may determine effectiveness of various advertisements using the information from the third party identifying conversions associated with different groups.
Abstract:
A camera system captures images from a set of cameras to generate binocular panoramic views of an environment. The cameras are oriented in the camera system to maximize the minimum number of cameras viewing a set of randomized test points. To calibrate the system, matching features between images are identified and used to estimate three-dimensional points external to the camera system. Calibration parameters are modified to improve the three-dimensional point estimates. When images are captured, a pipeline generates a depth map for each camera using reprojected views from adjacent cameras and an image pyramid that includes individual pixel depth refinement and filtering between levels of the pyramid. The images may be used generate views of the environment from different perspectives (relative to the image capture location) by generating depth surfaces corresponding to the depth maps and blending the depth surfaces.
Abstract:
A parallax viewer system allows 3D content, such as 360 degree 3D panoramas or other 3D environments, to be viewed by a user through a traditional 2D screen. A parallax viewer system operating on a user device can use a user-facing camera and a 2D screen to simulate a 3D environment for a user viewing the 2D screen. By changing the rendered view of the 3D environment as the user's head moves with respect to the screen, the parallax viewer system can provide many of the immersion benefits of a VR or 3D display using a traditional 2D display. In some implementations, the parallax viewer system can be calibrated to work in new situations (for example, on a new user device) by determining the relationship between the user-facing camera and the screen used to display the virtual environment.
Abstract:
Advertisement content associated with an entity having a physical location is selected for display to a user by a client device. A proximity of the client device to the physical location is determined. In response to determining the proximity between the client device and the physical location, an advertisement is generated that includes the advertisement content and an indicator of the physical location. The generated advertisement is sent to the client device for display to the user.
Abstract:
An online system identifies a conversion of an advertisement when a client device associated with an online system user is within a threshold distance of a physical location associated with the advertisement. The client device sends obfuscated information identifying the user associated with the client device and the location of the client device to an independent third party when the client device is within a threshold distance of the physical location associated with the advertisement. Using information from the client device, the third party requests information from the online system identifying one or more groups including the user. Based on information identifying groups including various online system users, the third party generates information identifying conversions associated with different groups. The online system may determine effectiveness of various advertisements using the information from the third party identifying conversions associated with different groups.
Abstract:
A 360 video system can render 360 stereoscopic content based on a virtual environment using a standard GPU rendering pipeline. In some embodiments, in order to improve efficiency in generating 360 stereoscopic content, a vertex shift technique can be used to approximate multiple viewpoints in a single 360 stereoscopic eye view. When rendering the virtual environment, each triangle of the virtual environment can be shifted to represent the view from a viewpoint corresponding to that triangle. Using vertex shift techniques, a virtual environment can be rendered into a 360 stereoscopic eye view in one pass of a GPU rendering pipeline, according to some embodiments.
Abstract:
A camera system captures images from a set of cameras to generate binocular panoramic views of an environment. The cameras are oriented in the camera system to maximize the minimum number of cameras viewing a set of randomized test points. To calibrate the system, matching features between images are identified and used to estimate three-dimensional points external to the camera system. Calibration parameters are modified to improve the three-dimensional point estimates. When images are captured, a pipeline generates a depth map for each camera using reprojected views from adjacent cameras and an image pyramid that includes individual pixel depth refinement and filtering between levels of the pyramid. The images may be used generate views of the environment from different perspectives (relative to the image capture location) by generating depth surfaces corresponding to the depth maps and blending the depth surfaces.
Abstract:
A camera system captures images from a set of cameras to generate binocular panoramic views of an environment. The cameras are oriented in the camera system to maximize the minimum number of cameras viewing a set of randomized test points. To calibrate the system, matching features between images are identified and used to estimate three-dimensional points external to the camera system. Calibration parameters are modified to improve the three-dimensional point estimates. When images are captured, a pipeline generates a depth map for each camera using reprojected views from adjacent cameras and an image pyramid that includes individual pixel depth refinement and filtering between levels of the pyramid. The images may be used generate views of the environment from different perspectives (relative to the image capture location) by generating depth surfaces corresponding to the depth maps and blending the depth surfaces.