Abstract: An accurate camera pose is determined by pairing a first camera with a second camera in proximity to one another, and by developing a known spatial relationship between them. An image from the first camera and an image from the second camera are analyzed to determine corresponding features in both images, and a relative homography is calculated from these corresponding features. A relative parameter, such as a focal length or an extrinsic parameter is used to calculate a first camera's parameter based on a second camera's parameter and the relative homography.

Abstract: Embodiments of this invention relate to preprocessing video such that visual elements, such as advertisements, can be inserted at a later point in time when the video is requested by a user. In an embodiment, a video and target data is received. The video is analyzed to determine how a target moves and changes over time. Metadata describing how the target moves and changes over time is stored. When a user requests an on-demand video such as from a video sharing site, a visual element is inserted into the video. The visual elements move and change as defined in the stored metadata. In this way, advertisements can be inserted into on-demand video such as video from a video sharing website.

Abstract: Embodiments of this invention relate to preprocessing video such that visual elements, such as advertisements, can be inserted at a later point in time when the video is requested by a user. In an embodiment, a video and target data is received. The video is analyzed to determine how a target moves and changes over time. Metadata describing how the target moves and changes over time is stored. When a user requests an on-demand video such as from a video sharing site, a visual element is inserted into the video. The visual elements move and change as defined in the stored metadata. In this way, advertisements can be inserted into on-demand video such as video from a video sharing website.

Abstract: Embodiments of this invention relate to preprocessing video such that visual elements, such as advertisements, can be inserted at a later point in time when the video is requested by a user. In an embodiment, a video and target data is received. The video is analyzed to determine how a target moves and changes over time. Metadata describing how the target moves and changes over time is stored. When a user requests an on-demand video such as from a video sharing site, a visual element is inserted into the video. The visual elements move and change as defined in the stored metadata. In this way, advertisements can be inserted into on-demand video such as video from a video sharing website.

Abstract: An accurate camera pose is determined by pairing a first camera with a second camera in proximity to one another, and by developing a known spatial relationship between them. An image from the first camera and an image from the second camera are analyzed to determine corresponding features in both images, and a relative homography is calculated from these corresponding features. A relative parameter, such as a focal length or an extrinsic parameter is used to calculate a first camera's parameter based on a second camera's parameter and the relative homography.

Abstract: Embodiments of this invention relate to preprocessing video such that visual elements, such as advertisements, can be inserted at a later point in time when the video is requested by a user. In an embodiment, a video and target data is received. The video is analyzed to determine how a target moves and changes over time. Metadata describing how the target moves and changes over time is stored. When a user requests an on-demand video such as from a video sharing site, a visual element is inserted into the video. The visual elements move and change as defined in the stored metadata. In this way, advertisements can be inserted into on-demand video such as video from a video sharing website.

Abstract: A method is provided for deriving three-dimensional camera viewpoint information from a two-dimensional video image of a three-dimensional venue captured by a camera. The method includes the steps of identifying a two-dimensional geometric pattern in the two-dimensional video image, measuring the two-dimensional geometric pattern, and calculating the three-dimensional camera viewpoint information using the measurements of the two-dimensional geometric pattern. The two-dimensional geometric pattern may be an ellipse that corresponds to a circle in the three-dimensional venue, such as the center circle in a soccer field. The three-dimensional camera viewpoint information is provided to a tracking program, which uses the information to track the two-dimensional geometric pattern, or other objects, in subsequently-captured video images.

Abstract: A method and apparatus for using a driver program, executing on a host processor, to control the execution of programs on an auxiliary processor is described. The host processor and auxiliary processor share a storage device. In this storage device, the driver program stores a number of ordered execution control blocks, each containing location information of a corresponding program and execution information for controlling the execution flow on the auxiliary processor. An execution control program, executing on the auxiliary processor, processes the execution control blocks in the order maintained by the driver program. The auxiliary processor processes each execution control block and executes its' corresponding program, taking the action specified by the execution information. In this manner, the driver program indirectly controls the execution of programs on the auxiliary processor.

Abstract: A live video insertion system (LVIS) is disclosed that allows insertion of static or dynamic images into a live video broadcast in a realistic fashion on a real time basis. Initially, natural landmarks in a scene that are suitable for subsequent detection and tracking are selected. The landmarks are typically distributed throughout the entire scene, such as a ballpark or football stadium. The field of view of the camera at any instant is normally significantly smaller than the full scene that may be panned. The LVIS uses a combination of pattern recognition techniques and camera sensor data (e.g., pan, tilt, zoom, etc.) to locate, verify and track target data. Camera sensors are well suited for the searching requirements of an LVIS, while pattern recognition and landmark tracking techniques are better suited for the image tracking requirements of LVIS.

Abstract: Video data for two or more video streams are processed by a single background command list containing one or more background commands and a single foreground command list containing one or more foreground commands, where the foreground command list is implemented only during a vertical blanking interval of the display raster of a monitor. The processed video data are then displayed during the raster period. In a preferred embodiment, if the implementation of a background command has not completed by the start of a vertical blanking interval, the background command is interrupted, the state of the background command implementation is saved, the foreground command list is implemented, and then the interrupted background command is resumed. The background commands are preferably those that do not directly affect the display bitmaps, while the foreground commands are those that do directly affect the display bitmaps.

Abstract: First and second application programs run on a host processor. The driver loads, runs, and unloads a first microcode program of the first application program onto an Intel.RTM. i750.RTM. Pixel Processor. The driver then loads, runs, and unloads a second microcode program of the second application program onto the Intel.RTM. i750.RTM. Pixel Processor. The driver repeats these steps such that the first and second application programs run concurrently.