Abstract: In one example, the present disclosure describes a device, computer-readable medium, and method for inferring the presence of an occluded entity in a video captured via drone. For instance, in one example, a video is obtained. The video is captured by a drone monitoring a field of view of a scene. It is determined, based on a combination of statistical reasoning and contextual modeling of the video, that an occluded entity is likely to be present, but not entirely visible, in the field of view. A signal is sent to the drone to instruct the drone to adjust its orientation to make the occluded entity more visible.

Abstract: A method includes receiving, by a device, a first image of a scene and a second image of at least a portion of the scene. The method includes identifying a first plurality of features from the first image and comparing the first plurality of features to a second plurality of features from the second image to identify a common feature. The method includes determining a particular subset of pixels that corresponds to the common feature, the particular subset of pixels corresponding to a first subset of pixels of the first image and a second subset of pixels of the second image. The method also includes generating a first image quality estimate of the first image based on a comparison of a first degree of variation within the first subset of pixels and a second degree of variation within the second subset of pixels.

Abstract: Aspects of the subject disclosure may include, for example, facilitating access to a synopsis of a missed portion of a media program, identifying subject matter in a segment of content in the media program that has not been presented, determining a particular time for presenting the summary according to the subject matter in the segment of content, and initiating a presentation of the summary at the particular time. Other embodiments are disclosed.

Abstract: A method may include a processor training a machine learning model with a training data set, computing a data distribution of the training data set, processing a stream of new data to determine a likelihood of the new data from the data distribution that is computed, incrementing a counter when the likelihood of the new data is less than a first threshold, and retraining the machine learning model when the counter exceeds a second threshold.

Abstract: Aspects of the subject disclosure may include, for example, a system and method for selecting media content for a group of persons located at a venue. The system and method identify a plurality of viewers in a vicinity of a venue that has one or more display devices from location data and extracts a content viewing preference of each viewer from their profile data. The process includes aggregating the content viewing preference of each of the plurality of viewers to generate an aggregated content profile. Based on the aggregated content profile, a list of content is generated. Next, a first group of viewers approaching a viewing range of a display device are sensed by imaging data. Any conflicts between content viewing preferences of the first group of viewers are detected and resolved based on weighting the viewing preferences of the group. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, embodiments provisioning a neural network comprising a plurality of layers. Further embodiments include provisioning a plurality of Markov logic state machines among the plurality of layers of the neural network resulting in a machine learning application. Additional embodiments include training the machine learning application using historical network video traffic resulting in a trained machine learning application. Also, embodiments include receiving current network video traffic. Embodiments include provisioning network resources to route the current network video traffic according to the trained machine learning application. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, embodiments include receiving training data comprising historical states for network resources of a communication network. Further embodiments include generating a Riemannian geometry of the communication network according to the historical states for network resources. Each network resource is one of a vertex or an edge within the first Riemannian geometry. Additional embodiments include determining paths along the Riemannian geometry. Each path comprises at least one network resource. Also, embodiments include determining a velocity vector of each path according to at least one historical state. Embodiments include identifying a group of velocity vectors that generate a sub-geometry within the Riemannian geometry and provisioning a portion of network resources according to the group of velocity vectors and the sub-geometry. Other embodiments are disclosed.

Abstract: A method that incorporates teachings of the subject disclosure may include, for example, determining, by a system comprising a processor, more common features of a plurality of images according to similarity matrices indicating relative similarities between instances of common features occurring within multiple images of the plurality of images, defining, by the system, cluster groups associated with the more common features, where each cluster group comprises cluster images of the plurality of images, and where the more common features are present in each the cluster images, and performing, by the system, quality-based filtering on the cluster images to identify a target cluster image to represent the cluster images for each of the cluster groups. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, a method comprising obtaining, by a processing system including a processor, first and second models for a structure of an object, based respectively on ground-level and aerial observations of the object. Model parameters are determined for a three-dimensional (3D) third model of the object based on the first and second models; the determining comprises a transfer learning procedure. Data representing observations of the object is captured at an airborne unmanned aircraft system (UAS) operating at an altitude between that of the ground-level observations and the aerial observations. The method also comprises dynamically adjusting the third model in accordance with the operating altitude of the UAS; updating the adjusted third model in accordance with the data; and determining a 3D representation of the structure of the object, based on the updated adjusted third model. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, facilitating access to a synopsis of a missed portion of a media program, identifying subject matter in a segment of content in the media program that has not been presented, determining a particular time for presenting the summary according to the subject matter in the segment of content, and initiating a presentation of the summary at the particular time. Other embodiments are disclosed.

Abstract: A particular method includes receiving a set of images, where each image of the set of images is related to a common scene. The method also includes determining a first subset of pixels of a first image of the set, where the first subset of pixels corresponds to a second subset of pixels of a second image of the set. The method further includes generating a first image quality estimate of the first image based on a comparison of the first subset of pixels and the second subset of pixels.

Abstract: A method that incorporates teachings of the subject disclosure may include, for example, determining, by a system comprising a processor, more common features of a plurality of images according to similarity matrices indicating relative similarities between instances of common features occurring within multiple images of the plurality of images, defining, by the system, cluster groups associated with the more common features, where each cluster group comprises cluster images of the plurality of images, and where the more common features are present in each the cluster images, and performing, by the system, quality-based filtering on the cluster images to identify a target cluster image to represent the cluster images for each of the cluster groups. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, systems and methods aggregating video content and adjusting the aggregate video content according to a training model. The adjusted aggregate video content comprises a first subset of the images and does not comprise a second subset of the images. The first subset of the images is determined by the training model based on a plurality of categories corresponding to a plurality of events. The illustrative embodiments also include presenting the adjusted aggregate video content and receiving identifications for the first subset of the images in the aggregate video content. Further, the illustrative embodiments include adjusting the training model according to the identifications and providing the adjusted training model to a network device. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, a method for determining a first set of features in first images of first media content, generating a similarity score by processing the first set of features with a favorability model derived by identifying generative features and discriminative features of second media content that is favored by a viewer, and providing the similarity score to a network for predicting a response by the viewer to the first media content. Other embodiments are disclosed.

Abstract: A system that incorporates the subject disclosure may perform operations, for example, including determining a first correlation value between a behavior of a first audience member and a subject matter being presented at a display processor, wherein the subject matter being presented at the display is observable by the first audience member. A second correlation value is determined between the behavior of the first audience member and a characteristic of an environment in a vicinity of the display and the first audience member. Responsive to the second correlation value exceeding the first correlation value, a determination is made as to whether a present channel selection resulting from user input initiated by the first audience member is responsive to the characteristic of the environment influencing the audience member more than the subject matter being presented at the display. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, systems and methods aggregating video content and adjusting the aggregate video content according to a training model. The adjusted aggregate video content comprises a first subset of the images and does not comprise a second subset of the images. The first subset of the images is determined by the training model based on a plurality of categories corresponding to a plurality of events. The illustrative embodiments also include presenting the adjusted aggregate video content and receiving identifications for the first subset of the images in the aggregate video content. Further, the illustrative embodiments include adjusting the training model according to the identifications and providing the adjusted training model to a network device. Other embodiments are disclosed.

Abstract: A system, method, and computer program product for estimating human body pose are described. According to one aspect, a human figure silhouette is segmented from a depth image of a human actor. Contour points are sampled along the human figure silhouette. Inner Distance Shape Context (IDSC) descriptors of the sample contour points are determined and compared to IDSC descriptors of the feature points in an IDSC gallery for similarity. For each of the feature points, the sample contour point with the IDSC descriptor that is most similar to an IDSC of the feature point is identified as that feature point in the depth image. An estimated pose of a human model is estimated based on the detected feature points and kinematic constraints of the human model.

Abstract: Aspects of the subject disclosure may include, for example, a method for determining a first set of features in first images of first media content, generating a similarity score by processing the first set of features with a favorability model derived by identifying generative features and discriminative features of second media content that is favored by a viewer, and providing the similarity score to a network for predicting a response by the viewer to the first media content. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, a system and method for selecting media content for a group of persons located at a venue. The system and method identify a plurality of viewers in a vicinity of a venue that has one or more display devices from location data and extracts a content viewing preference of each viewer from their profile data. The process includes aggregating the content viewing preference of each of the plurality of viewers to generate an aggregated content profile. Based on the aggregated content profile, a list of content is generated. Next, a first group of viewers approaching a viewing range of a display device are sensed by imaging data. Any conflicts between content viewing preferences of the first group of viewers are detected and resolved based on weighting the viewing preferences of the group. Other embodiments are disclosed.

Abstract: Aspects of the subject disclosure may include, for example, systems and methods aggregating video content and adjusting the aggregate video content according to a training model. The adjusted aggregate video content comprises a first subset of the images and does not comprise a second subset of the images. The first subset of the images is determined by the training model based on a plurality of categories corresponding to a plurality of events. The illustrative embodiments also include presenting the adjusted aggregate video content and receiving identifications for the first subset of the images in the aggregate video content. Further, the illustrative embodiments include adjusting the training model according to the identifications and providing the adjusted training model to a network device. Other embodiments are disclosed.