839results about "Video data clustering/classification" patented technology

Techniques for printing stored or recorded information that may include information of one or more types and may be stored in one or more multimedia documents. A graphical user interface (GUI) is provided that facilitates printing of information stored by the one or more multimedia documents or portions of the information stored by the one or more multimedia documents.

A method for identifying object images in a display of a video signal, storing object identification information in an overlay directory, and extracting object information relevant to identified objects from respective information databases is disclosed. Overlay directories from several video signals and associated object information are assembled in an overlay server accessed by computers directly or through a telecommunication network. Object images are selected by users of authoring computers accessing the overlay server. An overlay directory for a video recording indicates temporal and spatial coordinates of images of selected objects, a network address of a video source providing the video recording, descriptions of the selected objects, and addresses of servers providing detailed information of the selected objects. The objects may include commercial products, services, or educational material. The disclosed method therefore may be used for advertising or general information dissemination purposes.

The disclosure provides methods and systems for interactive video content delivery. An example method comprises receiving a video content such as live television or video streaming. The method can run one or more machine-learning classifiers on video frames of the video content to create classification metadata corresponding to the machine-learning classifiers and one or more probability scores associated with the classification metadata. Furthermore, the method can create one or more interaction triggers based on a set of predetermined rules and optionally user profiles. The method can determine that a condition for triggering at least one of the triggers is met and triggers at least one of the actions with regard to the video content based on the determination, the classification metadata, and the probability scores. For example, the action can deliver additional information, present recommendations, automatically edit the video content, or control delivery of video content.

An exemplary virtual reality media system presents a field of view of an immersive virtual reality world on a display screen of a media player device associated with a user. The field of view includes content of the immersive virtual reality world and dynamically changes in response to user input provided by the user as the user experiences the immersive virtual reality world. Additionally, the virtual reality media system detects that a gaze of the user is directed for a predetermined amount of time at a gaze target included within the field of view. In response to the detection, the virtual reality media system presents an interactive user interface associated with the gaze target. In some examples, the interactive user interface is presented within the field of view together with the content of the immersive virtual reality world. Corresponding methods and systems are also described.

The invention relates to a video classification processing method and device, computer equipment and a storage medium. The method comprises the steps of obtaining a target video; extracting multi-modal data of the target video; performing classification prediction on the data of each mode to obtain probability vectors which respectively correspond to each mode and are used for representing the probability that the target video belongs to each preset category; combining the probability vectors corresponding to the modalities; and predicting the final category of the target video according to the combined vectors. According to the scheme, the accuracy of video classification can be improved.

The embodiment of the invention discloses a model training method, a video category detection method and device, an electronic device and a computer readable medium. An embodiment of the video category detection method comprises the steps of extracting a key frame of a target video, and generating a key frame sequence; inputting the key frame sequence into a feature extraction model to obtain a feature information sequence corresponding to the key frame sequence; and inputting the feature information sequence into a video category detection model to obtain a category detection result of the target video. According to the method and the device, the video category detection efficiency is improved.

A method and apparatus for retrieving information relevant to tracked objects appearing in a display of a video signal is disclosed. The method is performed by a viewing computer having stored thereon an augmented display tool. In response to a user requesting the video signal, a content directory storing content information relevant to the tracked objects is acquired from a video-overlay server. The augmented display tool causes the viewing computer to acquire and display the video signal and record a time measurement and spatial coordinates of each point selected by a viewer using a pointing device. The augmented display tool uses the content directory to find an object identifier corresponding to each selected point and extracts relevant information from a global object directory maintained at the video-overlay server.

The embodiment of the invention discloses a video classification method. The method comprises the steps of obtaining a key frame image from a target video; inputting the key frame image into an imagesearch engine to obtain description information of the key frame image, and determining a keyword group of the key frame image according to the description information; obtaining text content characteristics corresponding to the keyword group; and determining a video type label of the target video according to the text content characteristics. By adopting the method and the device, the text content characteristics of the target video can be determined on the basis of the key frame images in the plurality of frame images forming the target video and the corresponding description information inthe image search engine, so that the video type label of the target video is obtained, and the video classification efficiency is improved.

The invention relates to a content classification method and a device, a computer device and a storage medium. The method comprises the steps of obtaining a target feature vector corresponding to to-be-classified target content; obtaining a target classification model obtained by training, wherein the target classification model comprises a first classification model and a second classification model; inputting the target feature vector into a first classification model to obtain a first content category corresponding to the target content, the first content category being a content category corresponding to the first classification hierarchy; obtaining first category feature information corresponding to the first classification hierarchy; inputting the first category feature information and the target feature vector into a second classification model to obtain a second content category corresponding to the target content, the second content category being a content category corresponding to a second classification level, and the level of the second classification level being lower than the level of the first classification level; and taking the first content category and the second content category as classification results corresponding to the target content. The method can improve the content classification accuracy.

The invention relates to a double-stream video classification method and device based on a cross-modal attention mechanism. The method is different from a traditional double-flow method. Information of two modes (even more modes) is fused before a prediction result. Therefore, the method is more efficient and sufficient. Meanwhile, information interaction is carried out in the earlier stage, a single branch has important information of another branch in the later stage, the precision of the single branch is flush with or even exceeds that of a traditional double-flow method, and the parameterquantity of the single branch is much less than that of the traditional double-flow method. Compared with a non-local neural network, the attention module designed by the invention can be in a cross-mode state instead of only using an attention mechanism in a single mode, and the effect of the method provided by the invention is equivalent to that of the non-local neural network under the condition that the two modes are the same.

The present invention provides an easy-to-use centralized service for providing and using advanced video and audio browsing and tagging methods to create a revised and improved video media set and for enabling a user to auto-create a fixed media form of the so-edited and so-improved video. The present invention also enables a system that allows users to select varying degrees of automated creation of a fixed media form recording following editing and revision steps potentially involving synchronized tagging and commenting aspects. Systems and operational modes are provided for labeling and formatting the auto-generated fixed media data.

The invention relates to a video classification method, device and equipment, and a computer readable storage medium. The method comprises the steps of obtaining at least two key frame picture segments of a to-be-classified video; according to the time-space information of the at least two key frame picture segments, segment semantic vectors of the at least two key frame picture segments are acquired respectively; performing bidirectional association fusion on the at least two fragment semantic vectors to obtain a global to-be-classified semantic vector of the to-be-classified video; obtaininga prediction probability of the to-be-classified video according to the global to-be-classified semantic vector; wherein the prediction probability is used for determining a classification result ofthe to-be-classified video. By adopting the method, the video classification accuracy can be effectively improved.

The invention discloses a method for generating a video by extracting a multimedia material based on a template. The method comprises the following steps: acquiring multimedia materials, preprocessingthe multimedia materials, performing tagging processing, outputting tags of the multimedia materials, and clustering the multimedia materials and the corresponding tags according to a preset clustering rule to obtain a plurality of data sets; obtaining template configuration data input by a user, and establishing a video template according to the template configuration data and a preset initial template; and automatically performing a video generation task according to the template configuration data, extracting the multimedia material according to the template configuration data and the label, generating a video according to the template configuration data and the extracted multimedia material, and outputting the video. According to the invention, the user do not need to search, screen and confirm video materials, and video generation tasks can be automatically carried out according to user requirements, thereby reducing repeated operation processes of the users.

The embodiment of the invention provides a video collection generation method and device, electronic equipment and a storage medium. The method comprises the following steps: classifying according tothe content of a video frame to obtain a classification result; Determining a specific video segment containing predetermined content according to the classification result; And generating a video collection based on the specific video segment.

A computer-implemented method for managing video contents includes collecting a plurality of keywords related to a topic, the keywords being collected using at least one dynamic data source. One or more sub-topics of the topic are identified using the keywords collected. A topic node in a taxonomy of topics is built, the topic node including a topic identifier for the topic, a child topic identifier for the sub-topics identified, and a keyword section for one or more of the keywords collected. A plurality of videos is organized using the topic node built to assist a user in locating a video of interest.

Systems and methods for compressing and sharing geotagged video in accordance with embodiments of the invention are disclosed. One embodiment includes receiving a captured video sequence, where at least one geographic location is associated with the captured video sequence, selecting a segment of the captured video sequence, identifying a set of relevant video segments from a geotagged video database based on the at least one geotag associated with the captured video sequence, determining the video segment from the set of relevant video segments that is the best match by comparing the similarity of the content in the video segments to the content of the selected segment from the captured video sequence, encoding the selected segment, where the selected segment is encoded using predictions that include references to the video segment that is the best match, and storing the encoded video segment in the geotagged video database.

The invention discloses a video auditing method and device, an auditing server and a storage medium. The method comprises the following steps: inputting each key video frame in a to-be-audited video into a violation classification model, and obtaining a classification score of each key video frame under a corresponding violation sub-class through a binary classification module under different violation sub-classes in the violation classification model; for each violation sub-category, fusing the classification score of each key video frame under the violation sub-category to obtain a violationscore of the to-be-audited video under the violation sub-category; and determining violation category composition of the to-be-audited video according to the violation scores of the to-be-audited video under different violation sub-categories and a preset violation threshold. According to the technical scheme provided by the invention, multi-violation category judgment of the to-be-audited videois realized, the problem of misjudgment or missed judgment of violation sub-categories is avoided, the audit independence of the to-be-audited video under different violation sub-categories is ensured, and the comprehensiveness and accuracy of video audit are improved.