Abstract: Provided herein are methods and systems for improved accessibility scoring for content items. A predicted accessibility score may be based on a plurality of multimodal features present within a content item. The plurality of multimodal features may include video features (e.g., based on video/image analysis), audio features (e.g., based on audio analysis), text-based features (e.g., based on closed-captioning analysis), features indicated by metadata (e.g., duration, genre, etc.), a combination thereof, and/or the like. A predicted accessibility score for a content item may indicate how accessible the content item may be for persons who are visually impaired, hearing impaired, cognitively impaired, etc., as well as for persons who desire to view content that requires less visual attention and/or audio attention as the case may be.

Abstract: Particular embodiments provide a system to determine ad segments in a video asset to enable subsequent ad replacement in video programs. The system is included in a multiple service operator (MSO) system that broadcasts video programs via a broadcast schedule. The MSO may not know the location of the ad segments in the video asset. To determine the ad segments, the MSO uses a classifier to classify video program segments and advertisements in the video asset. The classifier may be integrated with an nDVR system. By integrating with the nDVR system, particular embodiments may determine user behavior information, such as trick play commands, from the nDVR system. The classifier may use the user behavior information to detect ad segments in the video asset. In one embodiment, the classifier may fuse outputs from different detectors to detect and validate ad segments in the video program.

Abstract: Systems, methods, and devices relating to determining an introduction portion in a video program are described herein. A method may determine first and second hard-matching pairs of video segments in first and second video content such that video fingerprints of the first hard-matching pair match and video fingerprints of the second hard-matching pair also match. The method may classify a third pair of video segments in the first and second video content, sequentially between the first and second hard-matching pairs, as a soft-matching pair of video segments of an introduction portion. The method may use the classification of the third pair of video segments as a soft-matching pair to determine a model configured to determine that a pair of video segments in two video content items are a soft-matching pair of video segments of an introduction portion.

Abstract: Systems and methods are described for processing video data. The system may predict an optimal bit rate for a video segment that satisfies a desired level of quality. The desired level of quality may be associated with a quality metric. The encoder may predict the bit rate using a machine learning model trained based on an analysis of features extracted from video segments encoded with known bit rates. The trained machine learning model may then predict the optimal bit rate for a given video segment that achieves or satisfies the desired level of quality for the quality metric. The video segment may then be encoded based on the predicted bit rate.

Abstract: Disclosed is a method of associating, at a secondary device, secondary media content with primary media content being output at a primary device. The method includes receiving, at the secondary device, first information based upon the primary content being output at the primary device, wherein the first information includes at least one of an audio and a visual signal, determining at the secondary device second information corresponding to the first information, receiving at the secondary device one or more portions of secondary media content that have been made available by a third device, determining at the secondary device whether one or more of the portions of the secondary media content match one or more portions of the second information, and taking at least one further action upon determining that there is a match.

Abstract: A method of detecting frames in a video that demarcate a pre-determined type of video segment within the video is provided. The method includes identifying visually distinctive candidate marker frames within the video, grouping the candidate marker frames into a plurality of groups based on visual similarity, computing a collective score for each of the groups based on temporal proximity of each of the candidate marker frames within the group to related events occurring within the video, and selecting at least one of the groups based on the collective proximity scores as marker frames that demarcate the pre-determined type of video segment. A video processing electronic device and at least one non-transitory computer readable storage medium having computer program instructions stored thereon for performing the method are also provided.

Abstract: A method for creating derivative advertisements is provided. The method includes identifying a visual element based on an analysis of a primary advertisement asset such that the visual element is representative of a branding message provided within the primary advertisement. The method also includes identifying a call-to-action (CTA) data element based on the analysis and creating a derivative advertisement asset having the visual element and CTA data element. A system and a non-transitory computer-readable storage medium containing software for executing the method are also provided.

Abstract: Methods and systems are provided for bitrate adaptation of a video asset to be streamed to a client device for playback. The method includes selecting a representation from a manifest which expresses a set of representations available for each chunk of the video asset and generating a dynamic manifest for the video asset in which the representation selected for the at least one chunk is recommended for streaming to the client device. The selection of the representation recommended for the chunk may be based on at least one of historic viewing behavior of previous viewers of the chunk, content analysis information for the chunk, a level of available network bandwidth, a level of available network storage, and data rate utilization information of network resources including current, average, peak, and minimum data rate of network resources.

Abstract: A method of detecting frames in a video that demarcate a pre-determined type of video segment within the video is provided. The method includes identifying visually distinctive candidate marker frames within the video, grouping the candidate marker frames into a plurality of groups based on visual similarity, computing a collective score for each of the groups based on temporal proximity of each of the candidate marker frames within the group to related events occurring within the video, and selecting at least one of the groups based on the collective proximity scores as marker frames that demarcate the pre-determined type of video segment. A video processing electronic device and at least one non-transitory computer readable storage medium having computer program instructions stored thereon for performing the method are also provided.

Abstract: A modality attention network (MAN) may analyze a content item (e.g., video, multimedia, live communications, etc.) and accurately determine (e.g., recognize, identify, predict, etc.) an expression, such as an emotion. The MAN is scalable and may include subnetworks pre-trained independently for a specific modality. Pre-training each subnetwork allows better learning of intra-modal dynamics and may produce a more accurate result for a specific task than traditional modality fusion techniques.

Abstract: Methods and systems are provided for bitrate adaptation of a video asset to be streamed to a client device for playback. The method includes selecting a representation from a manifest which expresses a set of representations available for each chunk of the video asset and generating a dynamic manifest for the video asset in which the representation selected for the at least one chunk is recommended for streaming to the client device. The selection of the representation recommended for the chunk may be based on at least one of historic viewing behavior of previous viewers of the chunk, content analysis information for the chunk, a level of available network bandwidth, a level of available network storage, and data rate utilization information of network resources including current, average, peak, and minimum data rate of network resources.

Abstract: Particular embodiments analyze logos found in a video program to determine video demarcations in the video program. For example, a video demarcation may be content that marks (“marker content”) a transition from a first video content type to a second video content type. Marker content may be used so the user knows that a transition is occurring. Particular embodiments analyze the logos found in a video program to determine the video demarcations in the video. The video is first analyzed to determine logos in the video program. Once these logos are determined, particular embodiments may re-analyze the video program to identify marker frames that include the marker content that signal the transitions to a different video content types. The marker frames may be determined without any prior knowledge of the marker content. Then, particular embodiments may use the marker frames to determine video segments.

Abstract: Systems and methods for determining the location of advertisements in multimedia assets are disclosed. A method includes obtaining a multimedia asset comprising at least one of video, audio and text, identifying segments of the multimedia asset, and for each segment, determining whether the respective segment represents an advertisement. For each segment that represents an advertisement, the method further includes extracting a predetermined number of fingerprints from the respective segment, submitting, to an advertisement database, the predetermined number of fingerprints, receiving, from the advertisement database, a count of the extracted fingerprints matching fingerprints previously stored in the advertisement database, and determining, based on the count, whether to add an indication that the respective segment is a new advertisement to the advertisement database.

Abstract: Particular embodiments provide a system to determine ad segments in a video asset to enable subsequent ad replacement in video programs. The system is included in a multiple service operator (MSO) system that broadcasts video programs via a broadcast schedule. The MSO may not know the location of the ad segments in the video asset. To determine the ad segments, the MSO uses a classifier to classify video program segments and advertisements in the video asset. The classifier may be integrated with an nDVR system. By integrating with the nDVR system, particular embodiments may determine user behavior information, such as trick play commands, from the nDVR system. The classifier may use the user behavior information to detect ad segments in the video asset. In one embodiment, the classifier may fuse outputs from different detectors to detect and validate ad segments in the video program.

Abstract: Disclosed is a method of associating, at a secondary device, secondary media content with primary media content being output at a primary device. The method includes receiving, at the secondary device, first information based upon the primary content being output at the primary device, wherein the first information includes at least one of an audio and a visual signal, determining at the secondary device second information corresponding to the first information, receiving at the secondary device one or more portions of secondary media content that have been made available by a third device, determining at the secondary device whether one or more of the portions of the secondary media content match one or more portions of the second information, and taking at least one further action upon determining that there is a match.

Abstract: Particular embodiments provide a system to determine ad segments in a video asset to enable subsequent ad replacement in video programs. The system is included in a multiple service operator (MSO) system that broadcasts video programs via a broadcast schedule. The MSO may not know the location of the ad segments in the video asset. To determine the ad segments, the MSO uses a classifier to classify video program segments and advertisements in the video asset. The classifier may be integrated with an nDVR system. By integrating with the nDVR system, particular embodiments may determine user behavior information, such as trick play commands, from the nDVR system. The classifier may use the user behavior information to detect ad segments in the video asset. In one embodiment, the classifier may fuse outputs from different detectors to detect and validate ad segments in the video program.

Abstract: A first screen schedule for the linear delivery of video content is received and analyzed. The first screen schedule may include slots in which first screen advertisements could be displayed. Particular embodiments can then determine the feasibility of providing second screen advertisements based on the first screen schedule. Then, second screen advertisements that advertisers wish to display in conjunction with first screen advertisements are determined. When the linear delivery of the video content commences, particular embodiments monitor a messaging stream associated with the delivery of the video content. When an ad placement marker is encountered, an ad server may select a first screen advertisement. When the first screen advertisement is placed, the messaging stream identifies a first screen advertisement that is delivered. If the first screen advertisement has a corresponding second screen advertisement, particular embodiments output the second screen advertisement on a second screen environment.

Abstract: Disclosed is a method of associating, at a secondary device, secondary media content with primary media content being output at a primary device. The method includes receiving, at the secondary device, first information based upon the primary content being output at the primary device, wherein the first information includes at least one of an audio and a visual signal, determining at the secondary device second information corresponding to the first information, receiving at the secondary device one or more portions of secondary media content that have been made available by a third device, determining at the secondary device whether one or more of the portions of the secondary media content match one or more portions of the second information, and taking at least one further action upon determining that there is a match.

Abstract: A method of video segment detection within a transport stream of a video asset is provided. Boundaries of candidate video segments of interest (i.e., advertisements, sports highlights, news highlights, content summaries, etc.) within a video asset are detected with a media analysis detector and are separately detected based on statistical models generated from historic transport control event data collected from a population of viewers of the video asset. The above referenced information concerning the candidate video segments of interest is used to identify beginning and end boundaries of selected candidate video segments within the transport stream. A transcoder is provided with parameters corresponding to the selected candidate video segments and performs group of pictures (GOP) and chunk boundary alignment of chunks of the transport stream with the boundaries of the selected candidate video segments. A system and non-transitory computer-readable storage medium are also disclosed.

Abstract: A method is provided for detecting static graphical elements in a sequence of video frames that compares a selected frame in the sequence to each of a plurality of previous frames in the sequence to identify a graphical element that can be a logo. For each pair of frames compared, an absolute difference frame is determined by acquiring an absolute difference value between pixel values for corresponding pixels over at least a portion of the frames in the frame pair. A metric associated with each absolute difference frame is generated, which reflects a degree of dissimilarity. At least some of the absolute difference frames weighted in accordance with the metric associated therewith are summed to generate an accumulation difference frame such that pairs of frames that are more dissimilar have a greater weight. A static graphical element is then identified over a region of the accumulation difference frame in which pixel values satisfy specified criteria.