673results about "Video data indexing" patented technology

A multi-dimensional database and indexes and operations on the multi-dimensional database are described which include video search applications or other similar sequence or structure searches. Traversal indexes utilize highly discriminative information about images and video sequences or about object shapes. Global and local signatures around keypoints are used for compact and robust retrieval and discriminative information content of images or video sequences of interest. For other objects or structures relevant signature of pattern or structure are used for traversal indexes. Traversal indexes are stored in leaf nodes along with distance measures and occurrence of similar images in the database. During a sequence query, correlation scores are calculated for single frame, for frame sequence, and video clips, or for other objects or structures.

Fast metadata indexing and delivery for broadcast audio-visual (AV) programs by using template, segment-mark and bookmark on the visual spatio-temporal pattern of an AV program during indexing. The broadcasting time carried on a broadcast transport stream is used as a locator allowing direct access to a specific temporal position of a recorded AV program.

An apparatus for multi-bitrate content streaming includes a receiving module configured to capture media content, a streamlet module configured to segment the media content and generate a plurality of streamlets, and an encoding module configured to generate a set of streamlets. The system includes the apparatus, wherein the set of streamlets comprises a plurality of streamlets having identical time indices and durations, and each streamlet of the set of streamlets having a unique bitrate, and wherein the encoding module comprises a master module configured to assign an encoding job to one of a plurality of host computing modules in response to an encoding job completion bid. A method includes receiving media content, segmenting the media content and generating a plurality of streamlets, and generating a set of streamlets.

An apparatus for content item signature matching comprises a database (103) which has signatures for a plurality of content items. A likelihood processor (105) determines a match likelihood indication for the content items where the match likelihood indication is indicative of a likelihood of a match between the content item and an unknown signature. An interface (111) receives a query signature associated with a content item and in response a search processor (113) searches the database (103) for a matching signature to the query signature. The search processor (113) is operable to search the database in response to the match likelihood indication of the plurality of content items. In particular the database (103) may be ordered in order of decreasing probability of a match and the search processor (113) may search the database in this order. Hence, the probability of an early match is increased and the average search time is reduced.

The present invention relates to a system for generating a description record from multimedia information including, e.g., video data. A multimedia information input interface is used to receive multimedia information. A computer processor receives the multimedia information, processes the video information by performing video object extraction processing to generate video object descriptions from the video information, processes the generated video object descriptions by object hierarchy construction and extraction processing to generate video object hierarchy descriptions, and processes the generated video object descriptions by entity relation graph descriptions.

A method and system for description of synthetic audiovisual content makes it easier for humans, software components or devices to identify, manage, categorize, search, browse and retrieve such content. For instance, a user may wish to search for specific synthetic audiovisual objects in digital libraries, Internet web sites or broadcast media; such a search is enabled by the invention. Key characteristics of synthetic audiovisual content itself such as the underlying 2d or 3d models and parameters for animation of these models are used to describe it. More precisely, to represent features of synthetic audiovisual content, depending on the description scheme to be used, a number of descriptors are selected and assigned values. The description scheme instantiated with descriptor values is used to generate the description, which is then stored for actual use during query/search. Typically, a user, to search for a needed synthetic audiovisual content initiates a query that is passed on to a search engine that then retrieves the candidate content from one or more databases whose description closely matches the query criteria specified by the user.

Disclosed are systems and methods for reducing video communication bandwidth requirements of a video surveillance network camera system that includes network communication paths between network video cameras producing video streams of scenes observed by the network video cameras and content-aware computer networking devices analyzing by video analytics video visual content of the video streams to provide managed video representing, at specified quality levels, samples of the scenes observed. Distribution of the managed video consumes substantially less network bandwidth than would be consumed by delivery through network communication paths a video stream at the specified quality level in the absence of analysis by the video analytics.

A method and apparatus for deriving performance statistics from real time tracking of a sporting event. The method according to the present invention includes a step of obtaining a spatio-temporal trajectory corresponding to the motion of an athlete and based on real time tracking of the athlete. The trajectory is then broken down so that performance information corresponding to the motion of the athlete (such as speed, distance covered, acceleration, etc.) can be derived with respect to time. The information so obtained can be stored in a database or the like for later retrieval or can be used to graphically supplement a video broadcast of a sporting event. The apparatus includes a device for obtaining the trajectory, a computational device for obtaining the performance information based on the obtained trajectory, and a statistical device for compiling the performance information. In particular, the spatio-temporal trajectory may be obtained by an operation on a video image of the sporting event, in which motion regions in the image are identified, and feature points on the regions are tracked as they move, thereby defining feature paths. The feature paths, in turn, are associated in clusters, which clusters generally correspond to the motion of some portion of the athlete (e.g., arms, legs, etc.). The collective plurality of clusters corresponds with the motion of the athlete as a whole.

A video data recordable having integrated channel guides allowing a user to control recording and storage of television signals into personal channels for later playback and viewing. In the described embodiment, the user may specify criteria for recording of shows from an input source such as a broadcast signal and shows are then selected based on the user specified criteria and recorded for later playback. Storage of the shows may be organized into personal channels in order to facilitate later playback, e.g., the user may specify a channel of action movies, a channel of nature programming, a channel for sports, etc. The shows to be recorded may also have a predefined format which may be used to ease playback of recorded programming by allowing the user to easily locate and playback sections of programming of interest.

An intelligent video/audio observation and identification database system may define a security zone or group of zones. The system may identify vehicles and individuals entering or exiting the zone through image recognition of the vehicle or individual as compared to prerecorded information stored in a database. The system may alert security personnel as to warrants or other information discovered pertaining to the recognized vehicle or individual resulting from a database search. The system may compare images of a suspect vehicle, such as an undercarriage image, to standard vehicle images stored in the database and alert security personnel as to potential vehicle overloading or foreign objects detected, such as potential bombs. The system may further learn the standard times and locations of vehicles or individuals tracked by the system and alert security personnel upon deviation from standard activity.

A spherical content capture system captures spherical video content. A spherical video sharing platform enables users to share the captured spherical content and enables users to access spherical content shared by other users. In one embodiment, captured metadata or video/audio processing is used to identify content relevant to a particular user based on time and location information. The platform can then generate an output video from one or more shared spherical content files relevant to the user. The output video may include a non-spherical reduced field of view such as those commonly associated with conventional camera systems. Particularly, relevant sub-frames having a reduced field of view may be extracted from each frame of spherical video to generate an output video that tracks a particular individual or object of interest.

A spherical content capture system captures spherical video content. A spherical video sharing platform enables users to share the captured spherical content and enables users to access spherical content shared by other users. In one embodiment, captured metadata or video/audio processing is used to identify content relevant to a particular user based on time and location information. The platform can then generate an output video from one or more shared spherical content files relevant to the user. The output video may include anon-spherical reduced field of view such as those commonly associated with conventional camera systems. Particularly, relevant sub-frames having a reduced field of view may be extracted from each frame of spherical video o generate an output video that tracks a particular individual or object of interest.

Archival storage and retrieval of audio/video information is described. Audio and/or video information is digitized, compressed and stored in an intermediate archive format (IAF), which preserves the content at a high-enough quality for subsequent retrieval and conversion into various formats required at the time of use or distribution. A single capture operation is performed with ancillary metadata being added to facilitate subsequent searching, indexing and format conversion. Captured data content is catalogued and indexed at or subsequent to the creation of an IAF file that includes the archived information. The IAF includes a family of audio-video digital encoding formats based on public standards. The encoding format used in any particular application is determined at encoding time from information provided by the archive system user. At encoding, the particular encoding scheme is selected to optimize a tradeoff between storage constraints and end use quality requirements.

Stories are detected in multimedia data composed of concurrent streams for different modes such as audio, video and text and linked to related stories. First, time periods of uniformity in attributes of the streams serve as "building blocks" that are consolidated according to rules that are characteristic of the story to be detected. The attributes are then ranked by their respective reliabilities for detecting the story to be detected. An inter-attribute union of the time periods is cumulated attribute-to-attribute in an order that is based on the ranking. A buffered portion of the multimedia data that is delimited by the starting and ending times is retained in mass storage. The starting and ending times are indexed by characteristics of content of the portion to form a story segment which is maintained in a data structure with links to related story segments.

A rule storage unit 106 stores a plurality of pieces of specifying information each showing a feature of frames to be specified as start frames, and each corresponding to a different type of content. A program obtaining unit 104 obtains a content of which start frames are to be specified. An information obtaining unit 105 obtains type information showing the type of the obtained content. A selecting unit 107 obtains from the rule storage unit 106 a piece of specifying information corresponding to the obtained type information. A specifying unit 108 specifies, as the start frames, frames in the obtained content having the feature shown by the obtained piece of specifying information.

Systems and methods for assisting one or more users in the production of a media work from a plurality of recordings are disclosed. In an exemplary embodiment, digital versions of the recordings are stored on a file-storage server accessible over a communications network. A media-library server provides an organized way for users to access and view the digital versions, and to associate various data files with the digital versions, such as transcription files, logging files, and storyboard documents. The exemplary media library generates web pages that enable users to search and access the digital versions from an internet browser running on a conventional personal computer. The browser also enables users to view the digital versions with a browser-compatible media viewer.

A computer-assisted method for producing a space time summary for one or more original videos includes automatically recognizing a key element in an original video, extracting pixels related to the key element from a series of video frames of the original video, producing a video bit comprising a series of video frames comprising the pixels and audio information extracted from the original video, wherein at least one video frame of the video bit is formed by a subset of pixels of the corresponding video frame in the original video, automatically displaying a plurality of video bits in a user interface, wherein the plurality of video bits are extracted from one or more original videos, and provide a space time summary for the one or more original videos, and allowing two of the plurality of video bits to be played simultaneously with audio and motion in the user interface.

The invention discloses a remote damage assessment method and system based on distributed artificial intelligent image recognition. The method comprises the steps that an image collecting device collects external part images of a traffic accident vehicle and sends the external part images to a cloud platform; a vehicle-mounted sensor device acquires collision data information and uploads the collision data information to the cloud platform; an image processing module carries out preprocessing on the collected external part images of the traffic accident vehicle and stores the image preprocessing results into a vehicle damage assessment database; an image feature extraction module extracts features of the external part images according to the image preprocessing results, image matching is carried out on the obtained preprocessing external part image feature information of the traffic accident vehicle and an image base in the vehicle damage assessment database, and an external part damage assessment result of the traffic accident vehicle is obtained through a vehicle external part damage assessment model. By means of the method, on one hand, traffic accident vehicle external part damage assessment can be quickly carried out; on the other hand, by means of image and collision information step-by-step type damage assessment, the damage assessment faults can be reduced, and the precision and efficiency of settlement of claims can be improved.

Disclosed is a method and system for generating redacted video of a scene captured using a video camera and augmenting a manual review process of the redacted video of the scene. According to an exemplary embodiment, a video of the scene is redacted via probability-based analysis to detect and obscure privacy sensitive objects included in the captured video. A manual review process of the generated redacted video is augmented to use the object detection probability to enhance the reviewing video frame rate to expedite the manual review process.

A mobile terminal includes a display and controller configured to cause the display to display a playback screen of a first multimedia file and cause the display to display a first retrieval screen in response to receiving a first user input during the displaying of the playback screen of the first multimedia file, such that the first retrieval screen includes a plurality of thumbnail images respectively corresponding to one of a plurality of playback points in time on a per first time interval basis of the first multimedia file. The controller also causes the display to play the first multimedia file beginning at a playback point in time that corresponds to a selected one of the plurality of thumbnail images.

A conference information processing apparatus includes an extracting unit that extracts partial conference audio and/or visual information from conference audio and/or visual information. The partial conference audio and/or visual information has one or more physical events of a conference participant. The apparatus also has a providing unit that provides an index for the partial conference audio and/or visual information in accordance with a functional action abstracted from the one or more physical events.

Search tree structures with nodes that represent signatures derived from segments of video and audio content are used by systems to identify content and re-establish the correct alignment between video and audio content that have become disassociated with one another. The amount of storage needed to record data representing the tree structure can be reduced by replacing stored signature sets with signature pointers. The efficiency of searches in the tree structure can be improved by constructing and using partial tree structures.

An approach is provided for faceted browsing of video assets. A metadata file is parsed to categorize and rank attributes corresponding to a video asset, wherein the ranked attributes are based on a criterion. A set-top box provides attribute browsing to determine selection of the video asset.

The invention discloses a method for video matching in a video template library. A video template and an index of all feature codes corresponding to appearance positions of the feature codes in the video template are stored in the video template library, wherein the video template is obtained by connecting the feature codes of key frames in a time sequence. The method includes the steps that a query video is segmented, the key frames are selected from all query video segments, and the feature codes are extracted from all the key frames; the feature codes of all the key frames of all the query video segments are extracted, and the video template where the feature codes appear is searched for according to the index of the feature codes in the video template library; video template segments corresponding to the query video segments are found out in the video template according to the positions of the key frames in the query video segments, and whether the query video segments and the video template segments have a matching relation is judged through comparison based on the feature codes; the corresponding appearance position of the video template in the query video is obtained according to the matching relation between the query video segments and the video template segments. The method for video matching in the video template library improves video matching speed and robustness.

The invention provides a monitoring video file restoring method, which comprises the step of writing video file header information and also comprises the steps that the current video file is set to be in a writing state; video frame data is written into a media data object block, and media index information is periodically stored into a memory; after the writing of the video frame data is completed, the media index information is written into the video file tail, and in addition, the video file is set to be in a writing completion state; during video frame data abnormal writing interruption, video breakpoint information is recorded into a video breakpoint memory vault, and the current video file is set to be in a writing incompletion state; when video monitoring equipment is booted, if the existence of the video file in the writing incompletion state is detected, the video breakpoint information is read, the media index information of the video file is written into the breakpoint part of the video file, and the video file is set to be in the writing completion state. The monitoring video file restoring method has the advantages that after the power failure in the video recording process, the damaged video file is restored, and is recovered to the video several seconds earlier before the power failure, and the completeness of the video is ensured.

A method includes steps of indexing a media collection, searching an indexed library and browsing a set of candidate program segments. The step of indexing a media collection creates the indexed library based on a content of the media collection. The step of searching the indexed library identifies the set of candidate program segments based on a search criteria. The step of browsing the set of candidate program segments selects a segment for viewing.

The invention relates to a method for realizing the quick retrieval of mass videos. The method comprises the following steps: respectively extracting spatial feature vectors from all frame video images in a video stream of a video library to obtain video feature sequences; extracting key feature vectors from the spatial feature vectors; establishing a distributed storage index database according to the key feature vectors of all video files in the video library; extracting key feature vector sets of videos to be retrieved and extracting video index files of the videos to be retrieved; performing the video similarity retrieving in the distributed storage index database according to the video index files of the videos to be retrieved and outputting video retrieval results of the video files with the similarity larger than the preset value of the system. Through the adoption of the method with the structure, representative visual words are adopted to replace key frames, video information is completely represented, a large amount of redundant of video information does not exist, the video information is very compact, the retrieval speed is increased, and the method has mass data concurrent processing capacity, and is wider in application range.

The present invention is intended to make it possible to retrieve a desired scene from among numerous scenes by utilizing metadata that is recorded together with audiovisual data. Metadata describing each scene contained in audiovisual data recorded on a recording medium is analyzed and transformed into a metadata structure in which scenes are subordinated to each keyword. Moreover, transformed metadata is contained in a file, and then read in order to provide a scene retrieving feature that utilizes keywords.

A face recognition system is provided. The system includes a device configured to capture an input image of a subject. The system further includes a processor. The processor estimates, using a 3D Morphable Model (3DMM) conditioned Generative Adversarial Network, 3DMM coefficients for the subject of the input image. The subject varies from an ideal front pose. The processor produces, using an image generator, a synthetic frontal face image of the subject of the input image based on the input image and the 3DMM coefficients. An area spanning the frontal face of the subject is made larger in the synthetic image than in the input image. The processor provides, using a discriminator, a decision indicative of whether the subject of the synthetic image is an actual person. The processor provides, using a face recognition engine, an identity of the subject in the input image based on the synthetic and input images.

An imaging device detects a face of a subject from an image in response to inputting of the image containing the subject, and generates face data related to the face. The imaging device generates face data management information managing the face data and controls recording of the input image, the generated face data and the face data management information on a recording unit with the input image mapped to the face data and the face data management information. The face data contains a plurality information components recorded in a predetermined recording order. The face data management information, in a data structure responsive to the recording order of the information components of the face data, contains a train of consecutively assigned bits. The information components are assigned predetermined flags in the recording order. Each flag represents the presence or absence of the information component corresponding to the flag in the face data.