611 results about "Video retrieval" patented technology

A video retrieval system is presented that allows a user to quickly and easily select and receive stories of interest from a video stream. The video retrieval system classifies stories and delivers samples of selected stories that match each user's current preference. The user's preferences may include particular broadcast networks, persons, story topics, keywords, and the like. Key frames of each selected story are sequentially displayed; when the user views a frame of interest, the user selects the story that is associated with the key frame for more detailed viewing. This invention is particularly well suited for targeted news retrieval. In a preferred embodiment, news stories are stored, and the selection of a news story for detailed viewing based on the associated key frames effects a playback of the selected news story. The principles of this invention also allows a user to effect a directed search of other types of broadcasts as well. For example, the user may initiate an automated scan that presents samples of broadcasts that conform to the user's current preferences, akin to directed channel-surfing.

This invention provides a video search system based on content including an analysis server of video, phone, caption and man-face, a merging analysis server, a video search server cluster and a search dispatch server, in which, the analysis server can analyze and pick up video content information from the video and generate related spelling pattern information by analyzing audio signals to set up video index by analyzing and describing video content word information, the merging analyzing system merges data got from video analysis, phone analysis and caption analysis to generate index information, the video index server cluster stores the video content and the structure by the analysis system, the search system can enquires various information and feed back the result to the search dispatch server, which receives the research request of the customer end to analyze and transmit the request to the research cluster and generate ordered enquiry result.

A method and apparatus for video retrieval and cueing that automatically detects human faces in the video and identifies face-specific video frames so as to allow retrieval and viewing of person-specific video segments. In one embodiment, the method locates human faces in the video, stores the time stamps associated with each face, displays a single image associated with each face, matches each face against a database, computes face locations with respect to a common 3D coordinate system, and provides a means of displaying: 1) information retrieved from the database associated with a selected person or people, 2) path of travel associated with a selected person or people 3) interaction graph of people in video, 4) video segments associated with each person and/or face. The method may also provide the ability to input and store text annotations associated with each person, face, and video segment, and the ability to enroll and remove people from database. The videos of non-human objects may be processed in a similar manner. Because of the rules governing abstracts, this abstract should not be used to construe the claims.

An intelligent video retrieval system and method based on target characteristic and alarm behavior relate to an intelligent video retrieval system and method. The system comprises a trajectory tracking module, a feature extraction module, a feature storage module, a search engine module and a retrieval terminal, wherein the trajectory tracking module extracts the motion trajectory and alarm behavior of a target input in the video; the feature extraction module extracts the features of the input image such as color, texture, shape and category; the feature storage module stores the motion trajectory, the alarm behavior, the target characteristic and the related information of the video in a feature database; the search engine module analyzes the retrieval conditions input by a user, performs video retrieval based on characteristic and behavior in the database and returns the video clip satisfying the conditions; and the retrieval terminal, namely client software receives the retrieval conditions input by the user and displays the retrieval result after retrieval. Compared with the prior art, by using the retrieval system, the user can customize retrieval conditions more accurately, video retrieval based on target characteristic and alarm behavior can be performed; and the retrieval result of the features generated by real-time analysis can be fast obtained.

The invention discloses a video classification method based on a mixed convolution and attention mechanism, which solves the problems of complex calculation and low accuracy of the prior art. The method comprises the following steps of: selecting a video classification data set; Segmented sampling of input video; Preprocessing two video segments; Constructing hybrid convolution neural network model; The video mixed convolution feature map is obtained in the direction of temporal dimension. Video Attention Feature Map Obtained by Attention Mechanism Operation; Obtaining a video attention descriptor; Training the end-to-end entire video classification model; Test Video to be Categorized. The invention directly obtains mixed convolution characteristic maps for different video segments, Compared with the method of obtaining optical flow features, the method of obtaining optical flow features reduces the computational burden and improves the speed, introduces the attention mechanism betweendifferent video segments, describes the relationship between different video segments and improves the accuracy and robustness, and is used for video retrieval, video tagging, human-computer interaction, behavior recognition, event detection and anomaly detection.

This invention relates to a video content analysis system including: an input interface, a task dispatch module, a video stream obtaining module, a video content analysis module, a check module for video analysis result, an output interface and a UI. The system flow includes: the system receives the video analysis instruction from the network or the local, the task dispatch module decides the execution sequence of the task, the video flow obtaining module decodes the being analyzed video and the content analysis module analyzes and picks up lens information, the lens key frame information, the scene information, key frame information, its image information and man-face informatioin to be checked according to the need to be stored in the mode of XML file to be uploaded to the video information database via the output interface, and the system is used in supporting video search service based on content.

The invention relates to a recommendation method and system of video text labels. The method comprises: obtaining field words from the web to replenish a field dictionary at regular intervals; according to the field words in the field dictionary, carrying out segmentation on text information of each video to generate a plurality of candidate keywords; labeling an attribute type of each candidate keyword; according to the attribute type of each candidate keyword, obtaining a comprehensive weight number of the candidate keyword; sorting weight numbers of all the candidate keywords in descending order, and choosing several candidate keywords with the comprehensive weight numbers in the top as the video text labels. The recommendation method and system of the video text labels can automatically generate a video text label list, accurately summarize content of video and contribute to application such as video retrieval and excavation of relative videos.

The invention discloses a video analysis and storage method, a system and a video detection method and a system, respectively saving the storage space of the video image and increasing the efficiency of the video search, wherein the analysis and storage method and the system analyze the video image and store the motion segment containing the motion target, while the video detection method and the system analyze the video image and store motion segment containing the motion target, and the related motion target information and the motion segment information, search the stored motion segment according to the information and the index term input by the users. The invention can be widely applied in the video processing system, particularly the video monitoring system.

The invention provides a face clustering based video categorization method and a retrieval method as well as systems thereof. The method comprises steps as follows: a video file is subjected to shot segmentation; a video shot is subjected to face detection to obtain a face picture and relevant information; a detected face picture is subjected to duplicate removal; facial feature extraction is performed on the face picture subjected to duplicate removal, and extracted facial features are stored into a face feature library; feature clustering is performed on the face picture; the category of each face stored in a face category library is manually annotated; face metadata are automatically generated according to the manually annotated face category and the relevant information of a face in the category, the face metadata are added into an original catalogued file, and a final video catalogued file is obtained. The video categorization method, the retrieval method and the systems have the advantages as follows: the video file is automatically, accurately and efficiently catalogued, and manpower is greatly saved; a face sample library is not required to be prefabricated; a video retrieval function based on a human name and a picture is provided, and the convenient and efficient retrieval advantages are realized.

Techniques for determining if two video signals match by extracting features from a first and second video signal, and cross-correlating the features thereby providing a cross-correlation score at each of a number of time lags, then determining a mean magnitude of a difference of average values at each of the lags, and finally determining the similarity score based on both the cross-correlation scores and the mean magnitude difference scores, and optionally then outputting an indication of a degree of match between the first and second video signals.

This invention provides a search system based on video program content characters including a search dispatch server used in analyzing and dispatching the search requests put forward by users to search in terms of a certain strategy to grade, arrange and merge the search result to feed them back to the user, an element data search server searching the element data of the video program based on the requirement of the server, a caption search server searching the XML file storing program caption texts, a video search cluster used in searching the character data of the video key frame, a phone search cluster used in searching the phonetic information of video programs including spelling series and spelling patterns and a search interface facing to users receiving request of users in terms of signed message format and feeding back the search result.

Video surveillance systems are used to monitor public places, streets, buildings, cities and other premises or surroundings and comprise a multitude of cameras which are monitoring relevant points in the surroundings. One of the problems of surveillance systems is, that the multitude of cameras produce a large amount of video data to be monitored, which requires a lot of human observers resulting in high personnel costs. A video retrieval system (1) connectable to a plurality of surveillance cameras (2) for monitoring a surroundings is proposed, the video retrieval system (1) comprising: displaying means (12) realized to display a graphical overall description (100, 130, 140) of the surroundings under surveillance, whereby the graphical overall description (100, 130, 140) comprises a plurality of monitored areas (110), whereby the areas are allocated to different surveillance cameras (2) and whereby the graphical overall description (100, 130, 140) is displayed in a single perspective view, and querying means (10) realized to formulate a trajectory-specific search (150, 160) query adapted to retrieve data about a moving object in the surroundings with a matching trajectory.

Embodiments of the invention relate to a method for video retrieval by providing a first audio video file, determining a first identifier of a first piece of music in the first audio video file, looking up for the first identifier first meta data in a music database, in which said first meta data are associated with said first identifier, providing second meta data of a second piece of music from the music database. The second piece of music is included in a second audio video file. Further the method is conducted by determining a similarity measure by comparing the first and second meta data, and providing the second audio video file or an identifier thereof depending on the similarity measure. Further embodiments relate to a server, a user device and a system and a computer program product for video retrieval.

Techniques for determining if two video signals match by extracting features from a first and second video signal, and cross-correlating the features thereby providing a cross-correlation score at each of a number of time lags, and finally determining the similarity score based on both the cross-correlation scores.

Video frames of original video data are sampled with arbitrary time interval and size, and thumbnail frames are obtained. As thumbnail information concerning these frames, information on frame number of the original video frame corresponding each of the thumbnail frames and size of each thumbnail frame are described. Further, scene change information on the original video frames or intra-frame frame change value information are described altogether as additional information, and temporal/spatial thumbnail meta-data is obtained. The meta-data is associated with original video data, and a database is constructed. Then, the meta-data is employed, thereby performing typical frame display of original video data or variable speed reproduction. In this manner, even with a device with its low CPU capability, typical frame display or variable speed reproduction is performed for compressed and encoded video data such as MPEG-2, and the contents of video is checked, and retrieval is easily performed.

A video retrieval apparatus and a video retrieval method capable of executing appropriately a video retrieval and executing an effective retrieval by easily designating a retrieved object are provided. A video retrieval apparatus, includes a moving region extracting portion 102 that extracts a stored object moving region in a video, a region dividing portion 104 that divides the stored object moving region into stored object block regions, a representative color deriving portion 105 that derives representative colors of respective stored object block regions constituting the stored object moving region, a DB 106 that stores respective representative colors of respective stored object block regions constituting the stored object moving region, a retrieved object region designating portion 108 that extracts a retrieved object region in a video, a region dividing portion 109 that divides the retrieved object region into retrieved object block regions, a representative color calculating portion 110 that derives representative colors of respective retrieved object block regions constituting the retrieved object region, and a comparing portion 111 that compares respective representative colors of the stored object block regions with respective representative colors of the retrieved object block regions, and outputs compared results.

The present invention discloses a video big data rapid searching method and system constrained by abnormal behavior early-warning information, wherein a massive video monitoring data clue investigation range is constrained by using abnormal behavior early-warning information sent out by an intelligent monitoring camera. The method comprises: receiving abnormal behavior early-warning information of a front-end intelligent camera, and establishing an abnormal behavior library; establishing a correlation model table, wherein the correlation model table stores a correlation between a case or event type and an abnormal behavior type; and according to the correlation model table, searching the abnormal behavior library, and positioning a corresponding valuable clue of monitoring point The method and system disclosed by the technical scheme of the present invention significantly reduce the size of recording data called in the video searching process and greatly improve valuable clue discovery efficiency.

The invention relates to a generation method of a high-satisfaction video summary. Based on the characteristics of a video data structure, a shot cluster-based video summary system is designed and realized in the invention. The video summary system has the following main functions: coding, decoding and playing of various video files, shot boundary detection, key frame extraction, shot cluster, generation of both static video summary and user input time dynamic summary, and the like. The generation method is suitable for various application occasions, such as multimedia file management, video retrieval, and film and television database construction.

A query is received. The query may be an object containing temporal information. A query model including static and temporal components is then determined for the object. A weighting for static and temporal components is also determined. The query model is then compared with one or more search models. The search models also include static and temporal components. Search results are then determined based on the comparison. In one embodiment, the comparison may compare the static and temporal components of the query model and the search model. A weighting of the differences between the static and temporal components may be used to determine the ranking for the search results.

The invention discloses a video retrieval method and a video retrieval system. The video retrieval method comprises the following steps of: storing video clips into a video database, extracting corresponding affective characteristic vectors of each video clip, and storing the extracted affective characteristic vectors into a characteristic database; inputting the video clip to be retrieved, and extracting the affective characteristic vector of the video clip to be retrieved; matching the affective characteristic vector of the video clip to be retrieved with each affective characteristic vector in the characteristic database, calculating similarities between the affective characteristic vector of the video clip to be retrieved and each affective characteristic vector in the characteristic database, sequencing the similarities, and selecting a preset number of video clips with the greatest similarities; and outputting the selected video clips as retrieval results. By the technical scheme provided by the invention, the problems of affective characteristic extraction of the video clips and video content similarity matching can be solved, and the retrieval of video contents is realized by high-level semantics.

The invention provides a video retrieval method and a video retrieval system. The method includes: segmenting a video into a plurality of video clips with independent contents; obtaining a subject term of the video; subjecting each video clip to text labeling according to the subject term, making a video abstract of each video clip, setting up a semantic content index of the video according to the text label and the video abstract, and quickly browsing and retrieving the video content according to the semantic content index. By means of the video retrieval method and the video retrieval system, the video can be segmented into the plurality of video clips relatively independent in content to obtain the subject term of each video clip, the video is structured on basis of the subject terms, and the semantic content index of the video is set up, so that users can conveniently and quickly preview the video content to position interesting information, and user browsing and retrieving efficiency is improved.

The invention discloses a pedestrian re-identifying method based on coordination scale learning and belongs to the technical field of monitoring video retrieval. First, according to color and texture features of images in a marked training sample set L, scale learning is carried out, and covariance matrixes Mc and Mt in corresponding Mahalanobis distance are obtained; and checking targets are selected randomly, the Mc and the Mt are used for Mahalanobis distance measuring, a corresponding sorting result is obtained, positive samples and negative samples are obtained, a new marked training sample set L is obtained, the Mc and the Mt are updated until an unmarked training sample set U is empty, a final marked sample set L* is obtained, the color and texture features are fused, an Mf is obtained, and a Mahalanobis distance function based on the Mf can be used for pedestrian re-identifying. Under a semi-supervised framework, the pedestrian re-identifying technology based on scale learning is studied, scale learning is carried out with the marked samples assisted by the unmarked samples, the requirement that practical video investigation application marked training samples are hard to obtain is met, and re-identifying performance under few marked samples can be effectively improved.

The invention discloses a method for generating a video navigation system automatically, which comprises: cutting a video into a plurality of pictures by using a double-threshold picture cutting algorithm; extracting the key frames of the pictures by using a leading clustering algorithm; clustering the pictures by using the leading clustering algorithm to form a picture group; and creating a scene by using the picture group. The method improves the structuralization degree of the videos greatly, makes a hierarchical directory of the videos, and facilitates the video content capture and browsing of users. The method can make a DVD directory conveniently and provide support for video retrieval technology. Today, when videos have become part of people's life, the method has great application value and prospect.

The invention discloses a video fingerprint retrieval method based on coarse and fine granularity. According to the method, a video fingerprint is subjected to retrieval comparison in different-layer and different granularity retrieving mode. A local sensitive Hash algorithm is used for carrying out Hash processing on all video fingerprints in a video fingerprint database and the video fingerprint to be retrieved, so that the video fingerprint most similar to the video fingerprint to be retrieved is found, and the coarse granularity finding is completed, and the preparation is made for effectively reducing the time expenditure of the video retrieval; during the fine granularity finding, through the fast video fingerprint matching mechanism improved based on a biological sequence comparison technology BLAST, whether the video fingerprint to be retrieved exists in the video fingerprint database or not is fast found, the video segment matching finding can also be carried out, and in addition, the affiliated complete videos of the video segments to be retrieved and the concrete occurring time positions of the video segments to be retrieved in the complete videos can be determined according to the features of the video fingerprint, so that the accuracy and the real-time performance of the video finding are ensured.

The invention discloses a video retrieval method for a video cloud storage system. The method comprises: adding feature information of a shooting object into a video frame; establishing a data block index and a frame index in a cloud storage management cluster and a video storage cluster of the video cloud storage system respectively to establish a multilayer index structure; and performing video retrieval successively for a camera ID, shooting time and the feature information of the shooting object, so that the purposes of quickly reducing a retrieval range in huge video data and accurately obtaining a retrieval result are achieved. Meanwhile, the content of texts, pictures, audios and videos and the like can be analyzed by utilizing various algorithms in the method provided by the method, so that the search in multiple diversified search conditions is realized. Compared with the prior art, the method has the advantages that a large amount of events of manually searching for required video content and tracking monitoring targets are reduced, and the efficiency of searching for videos in massive video data is greatly improved.

A video retrieval system is provided, that includes a set of servers, configured to retrieve a video sequence from a database and forward it to a requesting device responsive to a match between an input text and a caption for the video sequence. The servers are further configured to translate the video sequence into the caption by (A) applying a C3D to image frames of the video sequence to obtain therefor (i) intermediate feature representations across L convolutional layers and (ii) top-layer features, (B) producing a first word of the caption for the video sequence by applying the top-layer features to a LSTM, and (C) producing subsequent words of the caption by (i) dynamically performing spatiotemporal attention and layer attention using the representations to form a context vector, and (ii) applying the LSTM to the context vector, a previous word of the caption, and a hidden state of the LSTM.