Abstract: The disclosure provides an approach for mimicking human camera operation with an autonomous camera system. In one embodiment, camera planning is formulated as a supervised regression problem in which an automatic broadcasting application receives one video input captured by a human-operated camera and another video input captured by a stationary camera with a wider field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera and the human-operated camera, respectively, and learns a regressor which takes as input such feature vectors and outputs pan-tilt-zoom settings predictive of what the human camera operator would choose. The automatic broadcasting application may then apply the learned regressor on newly captured video to obtain planned pan-tilt-zoom settings and control an autonomous camera to achieve the planned settings to record videos which resemble the work of a human operator in similar situations.

Abstract: The disclosure provides an approach for predicting trajectories for real-time capture of video and object tracking, while adhering to smoothness constraints so that predictions are not excessively jittery. In one embodiment, a temporally consistent search and learn (TC-SEARN) algorithm is applied to train a regressor for camera planning. A automatic broadcasting application first receives video input captured by a human-operated camera and another video input captured by a stationary camera with a wide field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera input and human-operated camera input, respectively. The automatic broadcasting application further applies the TC-SEARN algorithm to learn a sequential regressor for predicting camera trajectories, based on the extracted feature vectors and pan-tilt-zoom states.

Abstract: There is provided a system including a memory and a processor configured to receive a first image depicting a first object and a second image depicting a second object, divide the first image into a first plurality of patches and the second image into a second plurality of patches, extract plurality of feature vectors from each of the first plurality of patches and a second plurality of feature vectors from the second plurality of patches, determine dissimilarities based on a plurality of patch metrics, each patch dissimilarity measure being a dissimilarity between corresponding patches of the first plurality of patches and the second plurality of patches, compute an image dissimilarity between the first image and the second image based on an aggregate of the plurality of patch dissimilarity measures, evaluate the image dissimilarity to determine a probability of whether the first object and the second object are the same.

Abstract: The disclosure provides an approach for mimicking human camera operation with an autonomous camera system. In one embodiment, camera planning is formulated as a supervised regression problem in which an automatic broadcasting application receives one video input captured by a human-operated camera and another video input captured by a stationary camera with a wider field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera and the human-operated camera, respectively, and learns a regressor which takes as input such feature vectors and outputs pan-tilt-zoom settings predictive of what the human camera operator would choose. The automatic broadcasting application may then apply the learned regressor on newly captured video to obtain planned pan-tilt-zoom settings and control an autonomous camera to achieve the planned settings to record videos which resemble the work of a human operator in similar situations.

Abstract: Approaches are described for discovering a formation associated with an agent group engaging in an activity over a window of time. A formation analysis system computes first and second results for an objective function based on first and second sets of role assignments for each agent in the agent group at first and second moments in time, respectively. The formation analysis system iterates by: replacing the first set of role assignments with the second set of role assignments, and determining whether completion criteria have been met based at least in part on comparing the first result with the second result. If the completion criteria have not been met, then the formation analysis system replaces the second set of role assignments with a third set of role assignments that associate each agent in the first agent group with a different role assignment in the third set of role assignments at a third moment in time.

Abstract: The disclosure provides an approach for mimicking human camera operation with an autonomous camera system. In one embodiment, camera planning is formulated as a supervised regression problem in which an automatic broadcasting application receives one video input captured by a human-operated camera and another video input captured by a stationary camera with a wider field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera and the human-operated camera, respectively, and learns a regressor which takes as input such feature vectors and outputs pan-tilt-zoom settings predictive of what the human camera operator would choose. The automatic broadcasting application may then apply the learned regressor on newly captured video to obtain planned pan-tilt-zoom settings and control an autonomous camera to achieve the planned settings to record videos which resemble the work of a human operator in similar situations.

Abstract: There is provided a system including a memory and a processor configured to receive a first image depicting a first object and a second image depicting a second object, divide the first image into a first plurality of patches and the second image into a second plurality of patches, extract plurality of feature vectors from each of the first plurality of patches and a second plurality of feature vectors from the second plurality of patches, determine dissimilarities based on a plurality of patch metrics, each patch dissimilarity measure being a dissimilarity between corresponding patches of the first plurality of patches and the second plurality of patches, compute an image dissimilarity between the first image and the second image based on an aggregate of the plurality of patch dissimilarity measures, evaluate the image dissimilarity to determine a probability of whether the first object and the second object are the same.

Abstract: To generate a media presentation of a live event, a user interface is coupled to at least three cameras that share substantially the same vantage point. One of the cameras (e.g., a context camera) provides a context view of the event that is displayed on a screen of the user interface. The views of the other two cameras are superimposed onto the context view to define sub-portions that are visually demarcated within the context view. Based on user interaction, the user interface can switch between the cameras views and control the cameras to capture different portions of the context view. Based the image data captured by the views of the cameras within the context view, the user interface generates a media presentation that may be broadcast to multiple viewers.

Abstract: A device and method for receiving first detection information for a plurality of objects, the first detection information relating to a first characteristic of the objects, receiving second detection information for the objects, the second detection information relating to a second characteristic of the objects, determining first detections based upon the first detection information and second detections based upon the second detection information, formulating trellis graphs for the first and second detections, each trellis graph graphs including corresponding nodes at a plurality of time frames and determining a tracking of a selected one of the objects based upon a simultaneous shortest path for the selected object through both the first and second trellis graphs based upon a first path through the first trellis graph, a second path through the second trellis graph, and sidekick information.

Abstract: The disclosure provides an approach for mimicking human camera operation with an autonomous camera system. In one embodiment, camera planning is formulated as a supervised regression problem in which an automatic broadcasting application receives one video input captured by a human-operated camera and another video input captured by a stationary camera with a wider field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera and the human-operated camera, respectively, and learns a regressor which takes as input such feature vectors and outputs pan-tilt-zoom settings predictive of what the human camera operator would choose. The automatic broadcasting application may then apply the learned regressor on newly captured video to obtain planned pan-tilt-zoom settings and control an autonomous camera to achieve the planned settings to record videos which resemble the work of a human operator in similar situations.

Abstract: The disclosure provides an approach for predicting trajectories for real-time capture of video and object tracking, while adhering to smoothness constraints so that predictions are not excessively jittery. In one embodiment, a temporally consistent search and learn (TC-SEARN) algorithm is applied to train a regressor for camera planning. A automatic broadcasting application first receives video input captured by a human-operated camera and another video input captured by a stationary camera with a wide field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera input and human-operated camera input, respectively. The automatic broadcasting application further applies the TC-SEARN algorithm to learn a sequential regressor for predicting camera trajectories, based on the extracted feature vectors and pan-tilt-zoom states.

Abstract: There is provided a system for tracking objects. The system includes a processor and a memory for storing a plurality of sensory data frames. The processor determines a first hypothesized location for each of the objects in each of the plurality of sensory data frames. For each of the plurality of sensory data frames, the processor determines probabilities that the first hypothesized location of each of the objects in a sensory data frame of the plurality of sensory data frames is the same as the first hypothesized location of another object in an adjacent sensory data frame. The processor computes a first optimal trajectory for each of the objects using an algorithm based on the probabilities, checks the first optimal trajectory for each of the objects, and accepts or rejects the first optimal trajectory for each of the objects.

Abstract: Approaches are described for discovering a formation associated with an agent group engaging in an activity over a window of time. A formation analysis system computes first and second results for an objective function based on first and second sets of role assignments for each agent in the agent group at first and second moments in time, respectively. The formation analysis system iterates by: replacing the first set of role assignments with the second set of role assignments, and determining whether completion criteria have been met based at least in part on comparing the first result with the second result. If the completion criteria have not been met, then the formation analysis system replaces the second set of role assignments with a third set of role assignments that associate each agent in the first agent group with a different role assignment in the third set of role assignments at a third moment in time.

Abstract: There is provided a system for tracking objects. The system includes a processor and a memory for storing a plurality of sensory data frames. The processor determines a first hypothesized location for each of the objects in each of the plurality of sensory data frames. For each of the plurality of sensory data frames, the processor determines probabilities that the first hypothesized location of each of the objects in a sensory data frame of the plurality of sensory data frames is the same as the first hypothesized location of another object in an adjacent sensory data frame. The processor computes a first optimal trajectory for each of the objects using an algorithm based on the probabilities, checks the first optimal trajectory for each of the objects, and accepts or rejects the first optimal trajectory for each of the objects.

Abstract: To generate a media presentation of a live event, a user interface is coupled to at least three cameras that share substantially the same vantage point. One of the cameras (e.g., a context camera) provides a context view of the event that is displayed on a screen of the user interface. The views of the other two cameras are superimposed onto the context view to define sub-portions that are visually demarcated within the context view. In one embodiment, only one of views is visually demarcated in the context view at any given time. Based on user interaction, the user interface can switch between the cameras views and control the cameras to capture different portions of the context view. Based the image data captured by the views of the cameras within the context view, the user interface generates a media presentation that may be broadcast to multiple viewers.