WO2015061979A1 - 终端和视频文件管理方法 - Google Patents

终端和视频文件管理方法 Download PDF

Info

Publication number
WO2015061979A1
WO2015061979A1 PCT/CN2013/086207 CN2013086207W WO2015061979A1 WO 2015061979 A1 WO2015061979 A1 WO 2015061979A1 CN 2013086207 W CN2013086207 W CN 2013086207W WO 2015061979 A1 WO2015061979 A1 WO 2015061979A1
Authority
WO
WIPO (PCT)
Prior art keywords
frame image
video file
picture
time point
information
Prior art date
Application number
PCT/CN2013/086207
Other languages
English (en)
French (fr)
Inventor
黄小弟
Original Assignee
宇龙计算机通信科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 宇龙计算机通信科技(深圳)有限公司 filed Critical 宇龙计算机通信科技(深圳)有限公司
Priority to EP13896763.3A priority Critical patent/EP3065079A4/en
Priority to PCT/CN2013/086207 priority patent/WO2015061979A1/zh
Priority to CN201380068861.7A priority patent/CN104995639B/zh
Priority to US15/025,775 priority patent/US10229323B2/en
Publication of WO2015061979A1 publication Critical patent/WO2015061979A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/40Scenes; Scene-specific elements in video content
    • G06V20/46Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168Feature extraction; Face representation
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B27/00Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
    • G11B27/005Reproducing at a different information rate from the information rate of recording
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B27/00Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
    • G11B27/10Indexing; Addressing; Timing or synchronising; Measuring tape travel
    • G11B27/34Indicating arrangements 

Definitions

  • the present invention relates to the field of terminal technologies, and in particular, to a terminal and a video file management method. Background technique
  • the user drags the progress bar to locate the video playback by gesture.
  • the gesture stays on the progress bar, the screenshot of the location is displayed, and then the video continues to be played.
  • users want to watch the characters they are interested in they need to find them by dragging the player progress bar.
  • the drawback of the prior art is that after the video is played, when the user wants to find and watch the character play that he is interested in, he needs to drag the progress bar of the player while carefully staring at the screenshot of the video screen to see that the character he is interested in has It does not appear on the screen, it is cumbersome to operate, it takes time and effort, and it may be missed in the process of searching.
  • the invention is based on the above problems, and proposes a new technical solution, which can automatically select a video image that is of interest to the user in the video file according to the needs of the user, thereby avoiding the cumbersome operation of the user's own search and improving the user experience. .
  • the present invention provides a terminal, including: a picture recognition unit, performing identification processing on the acquired specified picture to extract feature information in the specified picture; and marking unit, according to the feature information, Searching for a frame image matching the specified picture in a preset manner in all frame images in the specified video file, and marking the found frame image; the processing unit is configured to: according to the received processing command, Said said marking unit marking The frame image performs a corresponding processing operation.
  • a frame image matching the specified picture can be found in the video file according to the feature information of the specified picture. For example, if the user wants to find all the pictures of the person A of interest in a certain video file, the user can obtain the picture of the character A, and the terminal will find all the pictures containing the character A in the video file according to the picture. In this way, the user can perform corresponding processing on the picture including the character A according to the individual's needs, such as directly playing all the pictures of the character A or synthesizing the picture of the character A into a video, playing or saving.
  • the specified picture includes, but is not limited to, a face image, and may be other images, such as a license plate image.
  • the user can extract the part of the video that is of interest to the user in the video file, thereby avoiding the user performing too many searching operations and improving the user experience.
  • the marking unit includes: an extracting subunit, configured to extract a partial frame image in all frame images of the specified video file according to a preset extraction step; And identifying the extracted partial frame image, identifying image information of the partial frame image; comparing the subunit, comparing the image information with the feature information, and marking the feature information a frame image corresponding to the matched image information; a determining subunit, determining whether the number of frame images spaced between adjacent two frame images in the marked frame image is equal to the preset extraction step size, in the determination result In the case of YES, the frame image between the adjacent two frame images is marked.
  • a part of the frame image may be extracted according to a preset extraction step in all frame images of the video file, and image information is identified for the extracted frame image, wherein the preset extraction step size may be If necessary, it is set by itself, for example, every 100 frames of images are extracted for recognition, and after the recognition is completed, it is determined whether there are 100 frames between adjacent two frame images that can match the specified picture, if the judgment result is Yes, it means that all the frame images between the adjacent two frame images can match the specified picture, and the 100 frames are also marked at the same time. This not only ensures the accuracy of the recognition result, but also reduces the processing capacity of the terminal and ensures the processing speed.
  • the match between the frame image and the specified picture is by comparing the feature information between the two.
  • the feature information is a facial feature
  • the terminal will recognize the facial feature information of the partial frame image extracted from the video file, and compare with the facial features of the specified picture, and the comparison result is consistent. , then the two match each other. If the specified picture is a license plate picture and the feature information is the license plate information, the terminal will recognize the license plate information in the partial frame image in the video file for matching.
  • the processing unit includes: a recording subunit, recording attribute information corresponding to the frame image, and performing corresponding processing operations on the frame image according to the attribute information, where
  • the attribute information includes time information and content information.
  • the terminal will record the attribute information of the frame image matching the specified picture, such as the time and content of the frame image, etc., so that the frame image can be played and synthesized according to the attribute information.
  • the recording subunit is further configured to: determine, according to the time information, a time point of the frame image on a time progress bar corresponding to the specified video file; and the processing unit The method further includes: a display subunit, configured to display the time progress bar, and highlighting the time point on the time progress bar.
  • the video clip of interest that is found by the user is highlighted on the time progress bar, so that the user can directly click to view the part of the video file that is of interest to him. It is not necessary to perform another search by yourself, which is convenient for the user's operation.
  • the display subunit is further configured to: select, according to the received selection instruction, a time point among all highlighted time points, and determine a time point adjacent to the selected time point. At a previous time point and/or a subsequent time point, the corresponding frame between the selected one time point and the previous time point or between the selected one time point and the latter time point is displayed in chronological order image.
  • the user can select a time point of interest in the found frame image according to personal needs, and can view the frame image before or after the time point by using the forward or backward option.
  • the processing unit further includes: a video generation subunit, and acquiring, according to the received video generation instruction, all frame patterns of the mark unit mark And recombining all the frame images into a new video file according to the attribute information corresponding to each frame image in all the frame images; and storing a subunit for storing the new video file .
  • the user can synthesize all the frame images of their own interest into a new video file, and the order of synthesis can be in the chronological order of the frame images, so that the user can view and store the new video files.
  • the picture recognition unit is further configured to: acquire the specified picture from the local and/or the Internet according to the received picture acquisition instruction.
  • the specified picture can be obtained from the local picture or from the network, providing a wide application space for the user.
  • the method further includes: a control unit, configured to: when receiving the function on or off command, control the picture recognition unit to start or stop performing recognition processing on the obtained specified picture.
  • the user can set whether to enable the above-mentioned picture recognition and other functions according to the needs of the individual, so that the necessary operations can be performed in a targeted manner to avoid unnecessary operations of the terminal.
  • a video file management method including: Step 202: Perform identification processing on the obtained specified image to extract feature information in the specified picture; Step 204, according to Characterizing information, searching for a frame image matching the specified picture in a preset manner in all frame images in the specified video file, and marking the found frame image; Step 206, according to the received processing The command performs a corresponding processing operation on the frame image of the tag.
  • a frame image matching the specified picture can be found in the video file according to the feature information of the specified picture. For example, if the user wants to find all the pictures of the person A of interest in a certain video file, the user can obtain the picture of the character A, and the terminal will find all the pictures containing the character A in the video file according to the picture. In this way, the user can perform corresponding processing on the picture including the character A according to the individual's needs, such as directly playing all the pictures of the character A or synthesizing the picture of the character A into a video, playing or saving.
  • the specified picture includes, but is not limited to, a face image, and may be other images, such as a license plate image.
  • the user can extract the part of the video that is of interest to the user in the video file, thereby avoiding the user performing too many searching operations and improving the user experience.
  • the step 204 includes: extracting a partial frame image in all frame images of the specified video file according to a preset extraction step; and identifying the extracted partial frame image And identifying the image information of the partial frame image; comparing the image information with the feature information, and marking a frame image corresponding to the image information that matches the feature information; determining the marked Whether the number of frame images spaced between adjacent two frame images in the frame image is equal to the preset extraction step size, and when the determination result is yes, marking the frame between the adjacent two frame images image.
  • a part of the frame image may be extracted according to a preset extraction step in all frame images of the video file, and image information is identified for the extracted frame image, wherein the preset extraction step size may be If necessary, it is set by itself, for example, every 100 frames of images are extracted for recognition, and after the recognition is completed, it is determined whether there are 100 frames between adjacent two frame images that can match the specified picture, if the judgment result is Yes, it means that all the frame images between the adjacent two frame images can match the specified picture, and the 100 frames are also marked at the same time. This not only ensures the accuracy of the recognition result, but also reduces the processing capacity of the terminal and ensures the processing speed.
  • the match between the frame image and the specified picture is achieved by comparing the feature information between the two. For example, if the specified picture is a character image, then the feature information is a facial feature, and the terminal will recognize the facial feature information of the partial frame image extracted from the video file, and compare with the facial features of the specified picture, and the comparison result is consistent. , then the two match each other. If the specified picture is a license plate picture and the feature information is the license plate information, the terminal will recognize the license plate information in the partial frame image in the video file for matching.
  • the step 206 includes: recording attribute information corresponding to the frame image, and performing corresponding processing operations on the frame image according to the attribute information, where the attribute information is Contains time information and content information.
  • the terminal will record the attribute letter of the frame image matching the specified picture.
  • Information such as the time and content of the frame image, so that the frame image can be played and synthesized according to the attribute information.
  • the step 206 further includes: determining, according to the time information, a time point of the frame image on a time progress bar corresponding to the specified video file; and displaying the time progress bar. And highlighting the point in time on the time progress bar.
  • the video clip of interest that is found by the user is highlighted on the time progress bar, so that the user can directly click to view the part of the video file that is of interest to him. It is not necessary to perform another search by yourself, which is convenient for the user's operation.
  • the step 206 further includes: selecting, according to the received selection instruction, a time point among all highlighted time points, and determining a previous one adjacent to the selected time point. At the time point and/or the subsequent time point, the frame images corresponding to the selected one time point and the previous time point or between the selected one time point and the latter time point are displayed in chronological order.
  • the user can select a time point of interest in the found frame image according to personal needs, and can view the frame image before or after the time point by using the forward or backward option.
  • the step 206 further includes: acquiring, according to the received video generation instruction, all the frame images of the mark, and according to the attribute information corresponding to each frame image in the all frame images, Reassembling all of the frame images into a new video file; and storing the new video file.
  • the user can synthesize all the frame images of their own interest into a new video file, and the order of synthesis can be in the chronological order of the frame images, so that the user can view and store the new video files.
  • the method further includes: acquiring the specified picture from the local and/or the Internet according to the received picture acquisition instruction.
  • the specified picture can be obtained from the local picture or from the online search, and provides a wide application space for the user.
  • the method further includes: starting or stopping to perform the identification processing on the acquired specified picture when receiving the function on or off command.
  • the user can set whether to enable the above-mentioned picture recognition and other functions according to the needs of the individual, so that the necessary operations can be performed in a targeted manner to avoid unnecessary operations of the terminal.
  • a program product stored on a non-transitory machine readable medium for video file management, the program product comprising a machine executable for causing a computer system to perform the following steps An instruction: performing identification processing on the obtained specified image to extract feature information in the specified picture; and searching for the specified picture according to a preset manner in all frame images in the specified video file according to the feature information Matching the frame image, and marking the found frame image; performing corresponding processing operations on the marked frame image according to the received processing command.
  • a non-volatile machine readable medium storing a program product for video file management, the program product comprising machine executable instructions for causing a computer system to perform the following steps: Obtaining a specified picture for performing identification processing to extract feature information in the specified picture; and searching, according to the feature information, a matching manner with the specified picture in a preset manner in all frame images in the specified video file a frame image, and marking the found frame image; performing corresponding processing operations on the marked frame image according to the received processing command.
  • a machine readable program the program causing a machine to execute the video file management method according to any one of the above aspects.
  • a storage medium storing a machine readable program, wherein the machine readable program causes the machine to execute the video file management method according to any one of the technical solutions described above.
  • FIG. 1 shows a block diagram of a terminal in accordance with an embodiment of the present invention
  • FIG. 3 shows a block diagram of a terminal in accordance with an embodiment of the present invention
  • FIG. 4 shows a flow chart of picture acquisition and recognition in accordance with an embodiment of the present invention
  • FIG. 5 shows a flow chart of tagging by a tag unit in accordance with an embodiment of the present invention
  • FIG. 1 shows a block diagram of a terminal in accordance with an embodiment of the present invention.
  • a terminal 100 includes: a picture recognition unit 102, performing identification processing on the acquired specified picture to extract feature information in the specified picture; and marking unit 104, according to The feature information is used to search for a frame image that matches the specified picture in a preset manner in all the frame images in the specified video file, and mark the searched frame image; the processing unit 106 is configured to The received processing command performs a corresponding processing operation on the frame image of the marking unit mark.
  • a frame image matching the specified picture can be found in the video file according to the feature information of the specified picture. For example, if the user wants to find all the pictures of the person A of interest in a certain video file, the user can obtain the picture of the character A, and the terminal will find all the pictures containing the character A in the video file according to the picture. In this way, the user can perform corresponding processing on the picture including the character A according to the individual's needs, such as directly playing all the pictures of the character A or synthesizing the picture of the character A into a video, playing or saving.
  • the specified picture includes but is not limited to a face image, and may be other images such as a license plate image.
  • the user can extract the part of the video that is of interest to the user in the video file, thereby avoiding the user performing too many searching operations and improving the user experience.
  • the marking unit 104 includes: an extracting sub-unit 1042, configured to extract a partial frame image in all frame images of the specified video file according to a preset extraction step; 1044, configured to identify the extracted partial frame image, and identify image information of the partial frame image; compare subunit 1046, compare the image information with the feature information, and mark the location a frame image corresponding to the image information matched by the feature information; a determining subunit 1048, determining whether the number of frame images spaced between adjacent two frame images in the marked frame image is equal to the preset extraction step Long, when the judgment result is YES, the frame image between the adjacent two frame images is marked.
  • an extracting sub-unit 1042 configured to extract a partial frame image in all frame images of the specified video file according to a preset extraction step
  • 1044 configured to identify the extracted partial frame image, and identify image information of the partial frame image
  • compare subunit 1046 compare the image information with the feature information, and mark the location a frame image corresponding to the image information matched by the feature information
  • a part of the frame image may be extracted according to a preset extraction step in all frame images of the video file, and image information is identified for the extracted frame image, wherein the preset extraction step size may be If necessary, it is set by itself, for example, every 100 frames of images are extracted for recognition, and after the recognition is completed, it is determined whether there are 100 frames between adjacent two frame images that can match the specified picture, if the judgment result is Yes, it means that all the frame images between the adjacent two frame images can match the specified picture, and the 100 frames are also marked at the same time. This not only ensures the accuracy of the recognition result, but also reduces the processing capacity of the terminal and ensures the processing speed.
  • the match between the frame image and the specified picture is achieved by comparing the feature information between the two. For example, if the specified picture is a character image, then the feature information is a facial feature, and the terminal will recognize the facial feature information of the partial frame image extracted from the video file, and compare with the facial features of the specified picture, and the comparison result is consistent. , then the two match each other. If the specified picture is a license plate picture and the feature information is the license plate information, the terminal will recognize the license plate information in the partial frame image in the video file for matching.
  • the processing unit 106 includes: a recording subunit 1062, recording attribute information corresponding to the frame image, and, according to the attribute information, the frame The image performs a corresponding processing operation, wherein the attribute information includes time information and content information.
  • the terminal will record the attribute information of the frame image matching the specified picture, such as the time and content of the frame image, etc., so that the frame image can be played and synthesized according to the attribute information.
  • the recording subunit 1062 is further configured to: determine, according to the time information, a time point of the frame image on a time progress bar corresponding to the specified video file; and the processing
  • the unit 106 further includes: a display subunit 1064 for displaying the time progress bar and highlighting the time point on the time progress bar.
  • the video clip of interest that is found by the user is highlighted on the time progress bar, so that the user can directly click to view the part of the video file that is of interest to him. It is not necessary to perform another search by yourself, which is convenient for the user's operation.
  • the display subunit 1064 is further configured to: select a time point among all highlighted time points according to the received selection instruction, and determine to be adjacent to the selected time point. a previous time point and/or a subsequent time point, displaying, in chronological order, a correspondence between the selected one time point and the previous time point or between the selected one time point and the latter time point Frame image.
  • the user can select a time point of interest in the found frame image according to personal needs, and can view the frame image before or after the time point by using the forward or backward option.
  • the processing unit 106 further includes: a video generation subunit 1066, acquiring all frame images of the mark unit mark according to the received video generation instruction, and following all the frames The attribute information corresponding to each frame image in the image is recombined into a new video file; and the storage subunit 1068 is configured to store the new video file.
  • the user can synthesize all the frame images of their own interest into a new video file, and the order of synthesis can be in the chronological order of the frame images, so that the user can view and store the new video files.
  • the picture identification unit 102 is further configured to: acquire the specified picture from a local and/or the Internet according to the received picture acquisition instruction.
  • the specified picture can be obtained from the local picture or from the network, providing a wide application space for the user.
  • the method further includes: a control unit 108, configured to: when receiving a function on or off command, control the picture recognition unit to start or stop performing recognition processing on the obtained specified picture.
  • the video file management method includes: Step 202: Perform identification processing on the obtained specified image to extract feature information in the specified picture; Step 204, according to Characterizing information, searching for a frame image matching the specified picture in a preset manner in all frame images in the specified video file, and marking the found frame image; Step 206, according to the received processing The command performs a corresponding processing operation on the frame image of the tag.
  • a frame image matching the specified picture can be found in the video file according to the feature information of the specified picture. For example, if the user wants to find all the pictures of the person A of interest in a certain video file, the user can obtain the picture of the character A, and the terminal will find all the pictures containing the character A in the video file according to the picture. In this way, the user can perform corresponding processing on the picture including the character A according to the individual's needs, such as directly playing all the pictures of the character A or synthesizing the picture of the character A into a video, playing or saving.
  • the specified picture includes, but is not limited to, a face image, and may be other images, such as a license plate image.
  • the user can extract the part of the video that is of interest to the user in the video file, thereby avoiding the user performing too many searching operations and improving the user experience.
  • the step 204 includes: extracting a partial frame image in all frame images of the specified video file according to a preset extraction step; and identifying the extracted partial frame image Identifying image information of the partial frame image; and Comparing the image information with the feature information, and marking a frame image corresponding to the image information matching the feature information; determining a frame image of the interval between two adjacent frame images in the marked frame image Whether the number is equal to the preset extraction step size, and when the determination result is YES, the frame image between the adjacent two frame images is marked.
  • a part of the frame image may be extracted according to a preset extraction step in all frame images of the video file, and image information is identified for the extracted frame image, wherein the preset extraction step size may be If necessary, it is set by itself, for example, every 100 frames of images are extracted for recognition, and after the recognition is completed, it is determined whether there are 100 frames between adjacent two frame images that can match the specified picture, if the judgment result is Yes, it means that all the frame images between the adjacent two frame images can match the specified picture, and the 100 frames are also marked at the same time. This not only ensures the accuracy of the recognition result, but also reduces the processing capacity of the terminal and ensures the processing speed.
  • the match between the frame image and the specified picture is achieved by comparing the feature information between the two. For example, if the specified picture is a character image, then the feature information is a facial feature, and the terminal will recognize the facial feature information of the partial frame image extracted from the video file, and compare with the facial features of the specified picture, and the comparison result is consistent. , then the two match each other. If the specified picture is a license plate picture and the feature information is the license plate information, the terminal will recognize the license plate information in the partial frame image in the video file for matching.
  • the step 206 includes: recording attribute information corresponding to the frame image, and performing corresponding processing operations on the frame image according to the attribute information, where the attribute information is Contains time information and content information.
  • the terminal will record the attribute information of the frame image matching the specified picture, such as the time and content of the frame image, etc., so that the frame image can be played and synthesized according to the attribute information.
  • the step 206 further includes: determining, according to the time information, a time point of the frame image on a time progress bar corresponding to the specified video file; and displaying the time progress bar. And highlighting the time on the time progress bar Point.
  • the video clip of interest that is found by the user is highlighted on the time progress bar, so that the user can directly click to view the part of the video file that is of interest to him. It is not necessary to perform another search by yourself, which is convenient for the user's operation.
  • the step 206 further includes: selecting, according to the received selection instruction, a time point among all highlighted time points, and determining a previous one adjacent to the selected time point. At the time point and/or the subsequent time point, the frame images corresponding to the selected one time point and the previous time point or between the selected one time point and the latter time point are displayed in chronological order.
  • the user can select a time point of interest in the found frame image according to personal needs, and can view the frame image before or after the time point by using the forward or backward option.
  • the step 206 further includes: acquiring, according to the received video generation instruction, all the frame images of the mark, and according to the attribute information corresponding to each frame image in the all frame images, Reassembling all of the frame images into a new video file; and storing the new video file.
  • the user can synthesize all the frame images of their own interest into a new video file, and the order of synthesis can be in the chronological order of the frame images, so that the user can view and store the new video files.
  • the method further includes: acquiring the specified picture from the local and/or the Internet according to the received picture acquisition instruction.
  • the specified picture can be obtained from the local picture or from the network, providing a wide application space for the user.
  • the method further includes: starting or stopping to perform the identification processing on the acquired specified picture when receiving the function on or off command.
  • the user can set whether to enable the above-mentioned picture recognition and other functions according to the needs of the individual, so that the necessary operations can be performed in a targeted manner to avoid unnecessary operations of the terminal.
  • the technical solution of the present invention will be described in detail below by taking a designated picture as a picture of a person.
  • FIG. 3 shows a block diagram of a terminal in accordance with an embodiment of the present invention.
  • the terminal 100 includes a picture acquiring unit 102, an identifying unit 104, and a processing unit 106. Each part is described in detail below.
  • the picture obtaining unit 102 is mainly responsible for acquiring a picture of a person of a character of interest to the user and extracting face recognition data.
  • Sources of image data include the user selecting a local character image or searching for a person's name via the web.
  • the identifying unit 104 by acquiring frame images (ie, frame images) in the video, matching and matching the frame images with the face data, finding out the frame images in which the characters of the user are interested, and then recording the time positions of the frames. And highlight it on the player progress bar with a certain color.
  • frame images ie, frame images
  • Processing unit 106 When the user searches for the character scene by clicking the forward or backward control button, the processing unit 106 continues to play according to the current playback time position, and locates the previous or next matching frame picture time position of the current time position. In addition, the user can select the clip synthesis to combine the clips of interest into one video and save them.
  • FIG. 4 shows a flow chart of picture acquisition and recognition in accordance with an embodiment of the present invention.
  • Step 402 The terminal starts the function of identifying the specified picture.
  • the user can set whether to enable the above-mentioned picture recognition function according to the needs of the individual, so that the necessary operations can be performed in a targeted manner to avoid unnecessary operations of the terminal.
  • Step 404 Prompt the user whether to obtain a character picture through the network.
  • the process proceeds to step 406. If the received selection result is NO, the process proceeds to step 408.
  • Step 406 Search for a picture by means of a person name search or the like.
  • Step 408 searching for a picture locally.
  • Step 410 Acquire a searched image, and collect face recognition data in the image.
  • FIG. 5 shows a flow chart of marking by a marking unit in accordance with an embodiment of the present invention.
  • Step 504 Perform the Nth frame picture and the face recognition data acquired in the image acquiring unit. Match.
  • step 506 it is determined whether the matching is successful. If the matching is successful, the process proceeds to step 508; otherwise, the process proceeds to step 510.
  • Step 508 Record and store the time position of the Nth frame picture.
  • N is incremented by 100. That is, the preset extraction step size is set to 100, and one frame image is extracted every 100 frames for recognition.
  • the preset extraction step size can also be set as needed, which not only ensures the accuracy of the recognition result, but also reduces the processing capacity of the terminal and ensures the processing speed.
  • Step 512 Determine whether N is greater than the total number of frames of the entire video. If yes, go to step 514; otherwise, go to step 504. If N is greater than the total number of frames of the entire video, it means that the video has been recognized, and the video can be processed.
  • Step 514 displaying the time position of all the records on the play progress bar, so that the user can perform further video processing operations.
  • the time interval between two adjacent time positions of the record is a preset time interval
  • the interval between two adjacent time positions is also displayed correspondingly.
  • the preset time The interval and the preset extraction step size correspond to each other, and the step size between the adjacent adjacent frame images is 100, indicating that the adjacent two frame images and the middle 100 frame images can be Matches the specified image, so its time position should also be displayed).
  • the video clips that are of interest to the user are highlighted on the time progress bar, and the user can directly click to view the part of the video file that he is interested in, without needing to
  • the search is convenient for the user's operation.
  • the user can also synthesize all the frame images of their own interest into a new video file, and the order of synthesis can be in the chronological order of the frame images, which is convenient for the user to view and store the new video files.
  • the user can click the "Function On 602" button to enable the image recognition function.
  • the user can see [local picture] on the interface. 604 and [Search Network Picture] 606 two menus, the user can select one of them to obtain the specified picture.
  • the user can obtain a picture of the person he wants to view or interested in, such as the user selecting [Local Picture] 604 and selecting Picture A from [Local Picture] 604.
  • the terminal recognizes the feature information in the picture A, and identifies the frame picture in the current video file to find a frame picture that matches the picture A.
  • the time progress bar 608 highlights the character play that the user is interested in, and adds two control buttons (the forward button 704 and the back button 702) and a video segment synthesis button (segment synthesis). Button 706).
  • the user can choose to drag the progress bar or click the forward button 704/back button 702 to quickly locate the video to play the character play that he is interested in.
  • the terminal can continue to play at the current playback time position and at the previous or next matching frame time position of the current time position.
  • the segment synthesis button 706 allows the user to combine the video segments of interest into one video file and retain them.
  • the technical solution of the present invention is described in detail above by taking a specific picture as a picture of a person, but the specific picture in the present invention includes but is not limited to a picture of a person, and may also be other images, such as a license plate image, etc., and if the specified picture is a license plate picture
  • the characteristic information is the license plate information
  • the terminal will recognize the license plate information in the partial frame image in the video file to perform corresponding matching, and the matching and processing process is similar to the above process, and details are not described herein again.
  • the technical solution of the present invention is described in detail with reference to the accompanying drawings.
  • the technical solution of the present invention can automatically select a video image that is of interest to the user in the video file according to the needs of the user, thereby avoiding the cumbersome operation of the user's own search and improving the user.
  • a program product stored on a non-transitory machine readable medium for video file management, the program product comprising a machine executable for causing a computer system to perform the following steps An instruction: performing identification processing on the obtained specified image to extract feature information in the specified picture; and searching for the specified picture according to a preset manner in all frame images in the specified video file according to the feature information Matching the frame image, and marking the found frame image; performing corresponding processing operations on the marked frame image according to the received processing command.
  • a nonvolatile machine readable medium stored a program product for video file management, the program product comprising machine executable instructions for causing a computer system to perform the following steps: performing recognition processing on the acquired specified image to extract feature information in the specified picture; And searching, according to the feature information, a frame image that matches the specified picture in a preset manner in all frame images in the specified video file, and marking the searched frame image; according to the received processing command And performing corresponding processing operations on the frame image of the mark.
  • a machine readable program the program causing a machine to execute the video file management method according to any one of the above aspects.
  • a storage medium storing a machine readable program, wherein the machine readable program causes the machine to execute the video file management method according to any one of the technical solutions described above.

Abstract

本发明提供了一种终端,包括:图片识别单元,对获取到的指定图片进行识别处理,以提取出指定图片中的特征信息;标记单元,根据特征信息,在指定视频文件中的所有帧图像中按照预设方式查找与指定图片相匹配的帧图像,并对查找出的帧图像进行标记;处理单元,用于根据接收到的处理命令,对标记单元标记的帧图像进行对应的处理操作。相应地,本发明还提供了一种视频文件管理方法。通过本发明的技术方案,可以根据用户的需要在视频文件中自动选择出用户感兴趣的视频画面,从而避免用户自己查找的繁琐操作,提升用户的使用体验。

Description

终端和视频文件管理方法
技术领域
本发明涉及终端技术领域, 具体而言, 涉及一种终端和一种视频文件 管理方法。 背景技术
当前的播放器在播放视频时, 用户通过手势拖动进度条去定位视频播 放, 手势停留在进度条上时便会显示该位置的画面截图, 然后继续播放视 频。 用户在想要观看自己感兴趣的角色戏份时, 需要通过拖动播放器进度 条自己去查找。
现有技术的缺陷在于, 视频播放后, 用户想要查找并观看自己感兴趣 的角色戏份时, 需要一边拖动播放器的进度条, 一边仔细盯着视频画面截 图, 查看自己感兴趣的角色有没有在画面中出现, 操作麻烦, 耗费时间精 力, 而且在查找的过程中可能会有所遗漏。
因此, 需要一种新的技术方案, 可以根据用户的需要在视频文件中自 动选择出用户感兴趣的视频画面, 从而避免用户自己查找的繁瑣操作, 提 升用户的使用体验。 发明内容
本发明正是基于上述问题, 提出了一种新的技术方案, 可以根据用户 的需要在视频文件中自动选择出用户感兴趣的视频画面, 从而避免用户自 己查找的繁瑣操作, 提升用户的使用体验。
有鉴于此, 本发明提出了一种终端, 包括: 图片识别单元, 对获取到 的指定图片进行识别处理, 以提取出所述指定图片中的特征信息; 标记单 元, 根据所述特征信息, 在指定视频文件中的所有帧图像中按照预设方式 查找与所述指定图片相匹配的帧图像, 并对查找出的所述帧图像进行标 记; 处理单元, 用于根据接收到的处理命令, 对所述标记单元标记的所述 帧图像进行对应的处理操作。
在该技术方案中, 可以根据指定图片的特征信息在视频文件中找到与 该指定图片相匹配的帧图像。 比如, 用户想要在某一视频文件中找出其感 兴趣的人物 A 的所有画面, 那么用户可以获取人物 A 的图片, 终端将根 据该图片在该视频文件中找到所有的含有人物 A的画面, 这样用户可以根 据个人的需要对包含人物 A的画面进行相应的处理, 如直接顺序播放人物 A的所有画面或者将人物 A的画面合成为一个视频, 播放或保存等。 当然 指定图片包括但是不限于人脸图像, 还可以是其他图像, 如车牌图像等。 通过该技术方案, 可以帮助用户在视频文件中抽取出自己感兴趣的那部分 视频, 从而避免了用户进行过多的查找操作, 提升了用户的使用体验。
在上述技术方案中, 优选地, 所述标记单元包括: 抽取子单元, 用于 根据预设的抽取步长在所述指定视频文件的所有帧图像中抽取出部分帧图 像; 识别子单元, 用于对抽取出的所述部分帧图像进行识别, 识别出所述 部分帧图像的图像信息; 比较子单元, 将所述图像信息和所述特征信息进 行对比, 并标记出与所述特征信息相匹配的图像信息所对应的帧图像; 判 断子单元, 判断已标记出的帧图像中相邻的两个帧图像之间间隔的帧图像 数是否等于所述预设的抽取步长, 在判断结果为是时, 标记出所述相邻的 两个帧图像之间的帧图像。
在该技术方案中, 由于一个视频文件中包含众多的帧图像, 如果对所 有的帧图像都进行信息识别, 那么终端的处理量势必很大, 并且连续的几 十帧图像或几百帧图像其差别并不是很大, 因此, 可以在视频文件的所有 帧图像中按照预设的抽取步长抽取出一部分帧图像, 对抽取出的帧图像进 行图像信息识别, 其中, 预设的抽取步长可以根据需要自行设置, 比如每 隔 100帧图像抽取一个帧图像进行识别, 而在识别完成之后再判断可与指 定图片匹配的相邻的两个帧图像之间是否间隔 100帧图像, 若判断结果为 是, 则说明相邻的两个帧图像之间的所有帧图像都可以与指定图片匹配, 此时也将这 100帧图像同时进行标记。 这样不但可以保证识别结果的准确 性, 而且可以减少终端的处理量, 保证处理的速度。
而帧图像和指定图片之间的匹配, 是通过比较两者之间的特征信息来 实现的。 比如, 指定图片是人物图像, 那么其特征信息就是面部特征, 终 端将识别出从视频文件中抽取出的部分帧图像的面部特征信息, 并与指定 图片的面部特征进行对比, 比较结果为一致的, 则说明两者相互匹配。 而 如果指定图片是车牌图片, 其特征信息就是车牌信息, 则终端会识别出视 频文件中的部分帧图像中的车牌信息进行匹配。
在上述技术方案中, 优选地, 所述处理单元包括: 记录子单元, 记录 所述帧图像对应的属性信息, 并根据所述属性信息, 对所述帧图像进行对 应的处理操作, 其中, 所述属性信息中包含时间信息和内容信息。
在该技术方案中, 终端将记录与指定图片相匹配的帧图像的属性信 息, 如帧图像的时间和内容等, 这样可以根据属性信息实现帧图像的播放 和合成等操作。
在上述技术方案中, 优选地, 所述记录子单元还用于: 根据所述时间 信息, 确定所述帧图像在所述指定视频文件对应的时间进度条上的时间 点; 以及所述处理单元还包括: 显示子单元, 用于显示所述时间进度条, 并在所述时间进度条上突出显示出所述时间点。
在该技术方案中, 在视频文件的播放界面, 会在时间进度条上突出显 示出为用户查找出的其感兴趣的视频片段, 这样用户可以直接点击查看到 视频文件中自己感兴趣的部分, 而不需要自己再另外进行查找, 方便了用 户的操作。
在上述技术方案中, 优选地, 所述显示子单元还用于: 根据接收到的 选择指令, 在被突出显示的所有时间点中选择一时间点, 并确定与选择的 一时间点相邻的前一时间点和 /或后一时间点, 按照时间顺序显示所选择 的一时间点和所述前一时间点之间或所选择的一时间点和所述后一时间点 之间所对应的帧图像。
在该技术方案中, 用户可以根据个人需要在查找出的帧图像中选择自 己感兴趣的时间点, 并可以通过前进或后退选项查看该时间点之前或之后 的帧图像。
在上述技术方案中, 优选地, 所述处理单元还包括: 视频生成子单 元, 根据接收到的视频生成指令, 获取所述标记单元标记的所有的帧图 像, 并按照所述所有的帧图像中每个帧图像对应的属性信息, 将所述所有 的帧图像重新组合成一个新的视频文件; 以及存储子单元, 用于存储所述 新的视频文件。
在该技术方案中, 用户可以将所有自己感兴趣的帧图像合成为一个新 的视频文件, 其合成的顺序可以是按照帧图像的时间顺序, 这样方便用户 对新的视频文件进行查看和存储。
在上述技术方案中, 优选地, 所述图片识别单元还用于: 根据接收到 的图片获取指令, 从本地和 /或互联网上获取所述指定图片。
在该技术方案中, 指定图片既可以从本地的图片中获取, 也可以从网 上搜索得到, 为用户提供了广泛的应用空间。
在上述技术方案中, 优选地, 还包括: 控制单元, 用于在接收到功能 开启或关闭命令时, 控制所述图片识别单元开始或停止对获取到的所述指 定图片进行识别处理。
在该技术方案中, 用户可以根据个人的需要设置是否开启上述的图片 识别等功能, 这样可以有针对性的进行必要操作, 避免终端的不必要操 作。
根据本发明的另一方面, 还提供了一种视频文件管理方法, 包括: 步 骤 202 , 对获取到的指定图片进行识别处理, 以提取出所述指定图片中的 特征信息; 步骤 204, 根据所述特征信息, 在指定视频文件中的所有帧图 像中按照预设方式查找与所述指定图片相匹配的帧图像, 并对查找出的所 述帧图像进行标记; 步骤 206, 根据接收到的处理命令, 对标记的所述帧 图像进行对应的处理操作。
在该技术方案中, 可以根据指定图片的特征信息在视频文件中找到与 该指定图片相匹配的帧图像。 比如, 用户想要在某一视频文件中找出其感 兴趣的人物 A 的所有画面, 那么用户可以获取人物 A 的图片, 终端将根 据该图片在该视频文件中找到所有的含有人物 A的画面, 这样用户可以根 据个人的需要对包含人物 A的画面进行相应的处理, 如直接顺序播放人物 A的所有画面或者将人物 A的画面合成为一个视频, 播放或保存等。 当然 指定图片包括但是不限于人脸图像, 还可以是其他图像, 如车牌图像等。 通过该技术方案, 可以帮助用户在视频文件中抽取出自己感兴趣的那部分 视频, 从而避免了用户进行过多的查找操作, 提升了用户的使用体验。
在上述技术方案中, 优选地, 所述步骤 204 包括: 根据预设的抽取步 长在所述指定视频文件的所有帧图像中抽取出部分帧图像; 对抽取出的所 述部分帧图像进行识别, 识别出所述部分帧图像的图像信息; 以及将所述 图像信息和所述特征信息进行对比, 并标记出与所述特征信息相匹配的图 像信息所对应的帧图像; 判断已标记出的帧图像中相邻的两个帧图像之间 间隔的帧图像数是否等于所述预设的抽取步长, 在判断结果为是时, 标记 出所述相邻的两个帧图像之间的帧图像。
在该技术方案中, 由于一个视频文件中包含众多的帧图像, 如果对所 有的帧图像都进行信息识别, 那么终端的处理量势必很大, 并且连续的几 十帧图像或几百帧图像其差别并不是很大, 因此, 可以在视频文件的所有 帧图像中按照预设的抽取步长抽取出一部分帧图像, 对抽取出的帧图像进 行图像信息识别, 其中, 预设的抽取步长可以根据需要自行设置, 比如每 隔 100帧图像抽取一个帧图像进行识别, 而在识别完成之后再判断可与指 定图片匹配的相邻的两个帧图像之间是否间隔 100帧图像, 若判断结果为 是, 则说明相邻的两个帧图像之间的所有帧图像都可以与指定图片匹配, 此时也将这 100帧图像同时进行标记。 这样不但可以保证识别结果的准确 性, 而且可以减少终端的处理量, 保证处理的速度。
而帧图像和指定图片之间的匹配, 是通过比较两者之间的特征信息来 实现的。 比如, 指定图片是人物图像, 那么其特征信息就是面部特征, 终 端将识别出从视频文件中抽取出的部分帧图像的面部特征信息, 并与指定 图片的面部特征进行对比, 比较结果为一致的, 则说明两者相互匹配。 而 如果指定图片是车牌图片, 其特征信息就是车牌信息, 则终端会识别出视 频文件中的部分帧图像中的车牌信息进行匹配。
在上述技术方案中, 优选地, 所述步骤 206 包括: 记录所述帧图像对 应的属性信息, 并根据所述属性信息, 对所述帧图像进行对应的处理操 作, 其中, 所述属性信息中包含时间信息和内容信息。
在该技术方案中, 终端将记录与指定图片相匹配的帧图像的属性信 息, 如帧图像的时间和内容等, 这样可以根据属性信息实现帧图像的播放 和合成等操作。
在上述技术方案中, 优选地, 所述步骤 206还包括: 根据所述时间信 息, 确定所述帧图像在所述指定视频文件对应的时间进度条上的时间点; 以及显示所述时间进度条, 并在所述时间进度条上突出显示出所述时间 点。
在该技术方案中, 在视频文件的播放界面, 会在时间进度条上突出显 示出为用户查找出的其感兴趣的视频片段, 这样用户可以直接点击查看到 视频文件中自己感兴趣的部分, 而不需要自己再另外进行查找, 方便了用 户的操作。
在上述技术方案中, 优选地, 所述步骤 206还包括: 根据接收到的选 择指令, 在被突出显示的所有时间点中选择一时间点, 并确定与选择的一 时间点相邻的前一时间点和 /或后一时间点, 按照时间顺序显示所选择的 一时间点和所述前一时间点之间或所选择的一时间点和所述后一时间点之 间所对应的帧图像。
在该技术方案中, 用户可以根据个人需要在查找出的帧图像中选择自 己感兴趣的时间点, 并可以通过前进或后退选项查看该时间点之前或之后 的帧图像。
在上述技术方案中, 优选地, 所述步骤 206还包括: 根据接收到的视 频生成指令, 获取标记的所有的帧图像, 并按照所述所有的帧图像中每个 帧图像对应的属性信息, 将所述所有的帧图像重新组合成一个新的视频文 件; 以及存储所述新的视频文件。
在该技术方案中, 用户可以将所有自己感兴趣的帧图像合成为一个新 的视频文件, 其合成的顺序可以是按照帧图像的时间顺序, 这样方便用户 对新的视频文件进行查看和存储。
在上述技术方案中, 优选地, 还包括: 根据接收到的图片获取指令, 从本地和 /或互联网上获取所述指定图片。
在该技术方案中, 指定图片既可以从本地的图片中获取, 也可以从网 上搜索得到, 为用户提供了广泛的应用空间。 在上述技术方案中, 优选地, 还包括: 在接收到功能开启或关闭命令 时, 开始或停止对获取到的所述指定图片进行识别处理。
在该技术方案中, 用户可以根据个人的需要设置是否开启上述的图片 识别等功能, 这样可以有针对性的进行必要操作, 避免终端的不必要操 作。
根据本发明的实施方式, 还提供了一种存储在非易失性机器可读介质 上的程序产品, 用于视频文件管理, 所述程序产品包括用于使计算机系统 执行以下步骤的机器可执行指令: 对获取到的指定图片进行识别处理, 以 提取出所述指定图片中的特征信息; 根据所述特征信息, 在指定视频文件 中的所有帧图像中按照预设方式查找与所述指定图片相匹配的帧图像, 并 对查找出的所述帧图像进行标记; 根据接收到的处理命令, 对标记的所述 帧图像进行对应的处理操作。
根据本发明的实施方式, 还提供了一种非易失机器可读介质, 存储有 用于视频文件管理的程序产品, 所述程序产品包括用于使计算机系统执行 以下步骤的机器可执行指令: 对获取到的指定图片进行识别处理, 以提取 出所述指定图片中的特征信息; 根据所述特征信息, 在指定视频文件中的 所有帧图像中按照预设方式查找与所述指定图片相匹配的帧图像, 并对查 找出的所述帧图像进行标记; 根据接收到的处理命令, 对标记的所述帧图 像进行对应的处理操作。
根据本发明的实施方式, 还提供了一种机器可读程序, 所述程序使机 器执行如上所述技术方案中任一所述的视频文件管理方法。
根据本发明的实施方式, 还提供了一种存储有机器可读程序的存储介 质, 其中, 所述机器可读程序使得机器执行如上所述技术方案中任一所述 的视频文件管理方法。
通过以上技术方案, 解决了用户需要拖动时间进度条去查找自己感兴 趣的视频图像的问题, 避免了用户的时间和精力的耗费, 提升了用户的使 用体验。 附图说明
图 1示出了根据本发明的实施例的终端的框图; 图 3示出了根据本发明的实施例的终端的结构图;
图 4示出了根据本发明的实施例的图片获取和识别的流程图; 图 5示出了根据本发明的实施例的标记单元进行标记的流程图; 意图。 具体实施方式
为了能够更清楚地理解本发明的上述目的、 特征和优点, 下面结合附 图和具体实施方式对本发明进行进一步的属性描述。 需要说明的是, 在不 沖突的情况下, 本申请的实施例及实施例中的特征可以相互组合。
在下面的描述中阐述了很多具体细节以便于充分理解本发明, 但是, 本发明还可以采用其他不同于在此描述的其他方式来实施, 因此, 本发明 的保护范围并不受下面公开的具体实施例的限制。
图 1示出了根据本发明的实施例的终端的框图。
如图 1 所示, 根据本发明的实施例的终端 100, 包括: 图片识别单元 102 , 对获取到的指定图片进行识别处理, 以提取出所述指定图片中的特 征信息; 标记单元 104 , 根据所述特征信息, 在指定视频文件中的所有帧 图像中按照预设方式查找与所述指定图片相匹配的帧图像, 并对查找出的 所述帧图像进行标记; 处理单元 106 , 用于根据接收到的处理命令, 对所 述标记单元标记的所述帧图像进行对应的处理操作。
在该技术方案中, 可以根据指定图片的特征信息在视频文件中找到与 该指定图片相匹配的帧图像。 比如, 用户想要在某一视频文件中找出其感 兴趣的人物 A 的所有画面, 那么用户可以获取人物 A 的图片, 终端将根 据该图片在该视频文件中找到所有的含有人物 A的画面, 这样用户可以根 据个人的需要对包含人物 A的画面进行相应的处理, 如直接顺序播放人物 A的所有画面或者将人物 A的画面合成为一个视频, 播放或保存等。 当然 指定图片包括但是不限于人脸图像, 还可以是其他图像, 如车牌图像等。 通过该技术方案, 可以帮助用户在视频文件中抽取出自己感兴趣的那部分 视频, 从而避免了用户进行过多的查找操作, 提升了用户的使用体验。
在上述技术方案中, 优选地, 所述标记单元 104 包括: 抽取子单元 1042 , 用于根据预设的抽取步长在所述指定视频文件的所有帧图像中抽取 出部分帧图像; 识别子单元 1044 , 用于对抽取出的所述部分帧图像进行 识别, 识别出所述部分帧图像的图像信息; 比较子单元 1046 , 将所述图 像信息和所述特征信息进行对比, 并标记出与所述特征信息相匹配的图像 信息所对应的帧图像; 判断子单元 1048 , 判断已标记出的帧图像中相邻 的两个帧图像之间间隔的帧图像数是否等于所述预设的抽取步长, 在判断 结果为是时, 标记出所述相邻的两个帧图像之间的帧图像。
在该技术方案中, 由于一个视频文件中包含众多的帧图像, 如果对所 有的帧图像都进行信息识别, 那么终端的处理量势必很大, 并且连续的几 十帧图像或几百帧图像其差别并不是很大, 因此, 可以在视频文件的所有 帧图像中按照预设的抽取步长抽取出一部分帧图像, 对抽取出的帧图像进 行图像信息识别, 其中, 预设的抽取步长可以根据需要自行设置, 比如每 隔 100帧图像抽取一个帧图像进行识别, 而在识别完成之后再判断可与指 定图片匹配的相邻的两个帧图像之间是否间隔 100帧图像, 若判断结果为 是, 则说明相邻的两个帧图像之间的所有帧图像都可以与指定图片匹配, 此时也将这 100帧图像同时进行标记。 这样不但可以保证识别结果的准确 性, 而且可以减少终端的处理量, 保证处理的速度。
而帧图像和指定图片之间的匹配, 是通过比较两者之间的特征信息来 实现的。 比如, 指定图片是人物图像, 那么其特征信息就是面部特征, 终 端将识别出从视频文件中抽取出的部分帧图像的面部特征信息, 并与指定 图片的面部特征进行对比, 比较结果为一致的, 则说明两者相互匹配。 而 如果指定图片是车牌图片, 其特征信息就是车牌信息, 则终端会识别出视 频文件中的部分帧图像中的车牌信息进行匹配。
在上述技术方案中, 优选地, 所述处理单元 106 包括: 记录子单元 1062 , 记录所述帧图像对应的属性信息, 并根据所述属性信息, 对所述帧 图像进行对应的处理操作, 其中, 所述属性信息中包含时间信息和内容信 息。
在该技术方案中, 终端将记录与指定图片相匹配的帧图像的属性信 息, 如帧图像的时间和内容等, 这样可以根据属性信息实现帧图像的播放 和合成等操作。
在上述技术方案中, 优选地, 所述记录子单元 1062 还用于: 根据所 述时间信息, 确定所述帧图像在所述指定视频文件对应的时间进度条上的 时间点; 以及所述处理单元 106 还包括: 显示子单元 1064 , 用于显示所 述时间进度条, 并在所述时间进度条上突出显示出所述时间点。
在该技术方案中, 在视频文件的播放界面, 会在时间进度条上突出显 示出为用户查找出的其感兴趣的视频片段, 这样用户可以直接点击查看到 视频文件中自己感兴趣的部分, 而不需要自己再另外进行查找, 方便了用 户的操作。
在上述技术方案中, 优选地, 所述显示子单元 1064 还用于: 根据接 收到的选择指令, 在被突出显示的所有时间点中选择一时间点, 并确定与 选择的一时间点相邻的前一时间点和 /或后一时间点, 按照时间顺序显示 所选择的一时间点和所述前一时间点之间或所选择的一时间点和所述后一 时间点之间所对应的帧图像。
在该技术方案中, 用户可以根据个人需要在查找出的帧图像中选择自 己感兴趣的时间点, 并可以通过前进或后退选项查看该时间点之前或之后 的帧图像。
在上述技术方案中, 优选地, 所述处理单元 106还包括: 视频生成子 单元 1066 , 根据接收到的视频生成指令, 获取所述标记单元标记的所有 的帧图像, 并按照所述所有的帧图像中每个帧图像对应的属性信息, 将所 述所有的帧图像重新组合成一个新的视频文件; 以及存储子单元 1068 , 用于存储所述新的视频文件。
在该技术方案中, 用户可以将所有自己感兴趣的帧图像合成为一个新 的视频文件, 其合成的顺序可以是按照帧图像的时间顺序, 这样方便用户 对新的视频文件进行查看和存储。 在上述技术方案中, 优选地, 所述图片识别单元 102还用于, 根据接 收到的图片获取指令, 从本地和 /或互联网上获取所述指定图片。
在该技术方案中, 指定图片既可以从本地的图片中获取, 也可以从网 上搜索得到, 为用户提供了广泛的应用空间。
在上述技术方案中, 优选地, 还包括: 控制单元 108 , 用于在接收到 功能开启或关闭命令时, 控制所述图片识别单元开始或停止对获取到的所 述指定图片进行识别处理。
在该技术方案中, 用户可以根据个人的需要设置是否开启上述的图片 识别等功能, 这样可以有针对性的进行必要操作, 避免终端的不必要操 作。 如图 2所示, 根据本发明的实施例的视频文件管理方法, 包括: 步骤 202 , 对获取到的指定图片进行识别处理, 以提取出所述指定图片中的特 征信息; 步骤 204 , 根据所述特征信息, 在指定视频文件中的所有帧图像 中按照预设方式查找与所述指定图片相匹配的帧图像, 并对查找出的所述 帧图像进行标记; 步骤 206 , 根据接收到的处理命令, 对标记的所述帧图 像进行对应的处理操作。
在该技术方案中, 可以根据指定图片的特征信息在视频文件中找到与 该指定图片相匹配的帧图像。 比如, 用户想要在某一视频文件中找出其感 兴趣的人物 A 的所有画面, 那么用户可以获取人物 A 的图片, 终端将根 据该图片在该视频文件中找到所有的含有人物 A的画面, 这样用户可以根 据个人的需要对包含人物 A的画面进行相应的处理, 如直接顺序播放人物 A的所有画面或者将人物 A的画面合成为一个视频, 播放或保存等。 当然 指定图片包括但是不限于人脸图像, 还可以是其他图像, 如车牌图像等。 通过该技术方案, 可以帮助用户在视频文件中抽取出自己感兴趣的那部分 视频, 从而避免了用户进行过多的查找操作, 提升了用户的使用体验。
在上述技术方案中, 优选地, 所述步骤 204 包括: 根据预设的抽取步 长在所述指定视频文件的所有帧图像中抽取出部分帧图像; 对抽取出的所 述部分帧图像进行识别, 识别出所述部分帧图像的图像信息; 以及将所述 图像信息和所述特征信息进行对比, 并标记出与所述特征信息相匹配的图 像信息所对应的帧图像; 判断已标记出的帧图像中相邻的两个帧图像之间 间隔的帧图像数是否等于所述预设的抽取步长, 在判断结果为是时, 标记 出所述相邻的两个帧图像之间的帧图像。
在该技术方案中, 由于一个视频文件中包含众多的帧图像, 如果对所 有的帧图像都进行信息识别, 那么终端的处理量势必很大, 并且连续的几 十帧图像或几百帧图像其差别并不是很大, 因此, 可以在视频文件的所有 帧图像中按照预设的抽取步长抽取出一部分帧图像, 对抽取出的帧图像进 行图像信息识别, 其中, 预设的抽取步长可以根据需要自行设置, 比如每 隔 100帧图像抽取一个帧图像进行识别, 而在识别完成之后再判断可与指 定图片匹配的相邻的两个帧图像之间是否间隔 100帧图像, 若判断结果为 是, 则说明相邻的两个帧图像之间的所有帧图像都可以与指定图片匹配, 此时也将这 100帧图像同时进行标记。 这样不但可以保证识别结果的准确 性, 而且可以减少终端的处理量, 保证处理的速度。
而帧图像和指定图片之间的匹配, 是通过比较两者之间的特征信息来 实现的。 比如, 指定图片是人物图像, 那么其特征信息就是面部特征, 终 端将识别出从视频文件中抽取出的部分帧图像的面部特征信息, 并与指定 图片的面部特征进行对比, 比较结果为一致的, 则说明两者相互匹配。 而 如果指定图片是车牌图片, 其特征信息就是车牌信息, 则终端会识别出视 频文件中的部分帧图像中的车牌信息进行匹配。
在上述技术方案中, 优选地, 所述步骤 206 包括: 记录所述帧图像对 应的属性信息, 并根据所述属性信息, 对所述帧图像进行对应的处理操 作, 其中, 所述属性信息中包含时间信息和内容信息。
在该技术方案中, 终端将记录与指定图片相匹配的帧图像的属性信 息, 如帧图像的时间和内容等, 这样可以根据属性信息实现帧图像的播放 和合成等操作。
在上述技术方案中, 优选地, 所述步骤 206还包括: 根据所述时间信 息, 确定所述帧图像在所述指定视频文件对应的时间进度条上的时间点; 以及显示所述时间进度条, 并在所述时间进度条上突出显示出所述时间 点。
在该技术方案中, 在视频文件的播放界面, 会在时间进度条上突出显 示出为用户查找出的其感兴趣的视频片段, 这样用户可以直接点击查看到 视频文件中自己感兴趣的部分, 而不需要自己再另外进行查找, 方便了用 户的操作。
在上述技术方案中, 优选地, 所述步骤 206还包括: 根据接收到的选 择指令, 在被突出显示的所有时间点中选择一时间点, 并确定与选择的一 时间点相邻的前一时间点和 /或后一时间点, 按照时间顺序显示所选择的 一时间点和所述前一时间点之间或所选择的一时间点和所述后一时间点之 间所对应的帧图像。
在该技术方案中, 用户可以根据个人需要在查找出的帧图像中选择自 己感兴趣的时间点, 并可以通过前进或后退选项查看该时间点之前或之后 的帧图像。
在上述技术方案中, 优选地, 所述步骤 206还包括: 根据接收到的视 频生成指令, 获取标记的所有的帧图像, 并按照所述所有的帧图像中每个 帧图像对应的属性信息, 将所述所有的帧图像重新组合成一个新的视频文 件; 以及存储所述新的视频文件。
在该技术方案中, 用户可以将所有自己感兴趣的帧图像合成为一个新 的视频文件, 其合成的顺序可以是按照帧图像的时间顺序, 这样方便用户 对新的视频文件进行查看和存储。
在上述技术方案中, 优选地, 还包括: 根据接收到的图片获取指令, 从本地和 /或互联网上获取所述指定图片。
在该技术方案中, 指定图片既可以从本地的图片中获取, 也可以从网 上搜索得到, 为用户提供了广泛的应用空间。
在上述技术方案中, 优选地, 还包括: 在接收到功能开启或关闭命令 时, 开始或停止对获取到的所述指定图片进行识别处理。
在该技术方案中, 用户可以根据个人的需要设置是否开启上述的图片 识别等功能, 这样可以有针对性的进行必要操作, 避免终端的不必要操 作。 下面以指定图片为人物图片为例, 详细说明本发明的技术方案。
图 3示出了根据本发明的实施例的终端的结构图。
如图 3 所示, 根据本发明的实施例的终端 100 , 包括图片获取单元 102 , 标识单元 104 , 处理单元 106, 下面详细说明各部分。
图片获取单元 102: 主要负责获取用户感兴趣角色的人物图片进而提 取人脸识别数据。 图片数据的来源包括用户选择本地人物图片或通过网络 搜索人名获取人物图片。
标识单元 104: 通过获取视频中的帧画面 (即帧图像) , 将这些帧画 面与人脸数据进行识别匹配, 找出那些出现用户所感兴趣角色的帧画面, 然后记录下这些帧画面的时间位置, 并通过一定的颜色在播放器进度条上 进行突出显示。
处理单元 106: 用户通过点击前进或后退控制按钮查找角色戏份时, 处理单元 106根据当前的播放时间位置, 定位到当前时间位置的前一个或 后一个匹配好的帧画面时间位置处继续播放。 此外用户可以选择片段合成 将兴趣片段合成一个视频并保存。
图 4示出了根据本发明的实施例的图片获取和识别的流程图。
如图 4所示, 根据本发明的实施例的图片识别过程的流程如下: 步骤 402 , 终端开启对指定图片进行识别的功能。 其中, 用户可以根 据个人的需要设置是否开启上述的图片识别等功能, 这样可以有针对性的 进行必要操作, 避免终端的不必要操作。
步骤 404 , 提示用户是否通过网络获取人物图片。 在接收到选择结果 为是时, 进入步骤 406, 在接收到的选择结果为否时, 进入步骤 408。
步骤 406 , 通过人名搜索等方式搜索图片。
步骤 408 , 在本地搜索图片。
步骤 410 , 获取搜索到的图片, 并采集图片中的人脸识别数据。
图 5示出了根据本发明的实施例的标记单元进行标记的流程图。
如图 5所示, 根据本发明的实施例的标记单元进行标记的流程如下: 步骤 502 , 初始化, 从视频的第 N帧画面开始, N=l。
步骤 504, 将第 N帧画面与图像获取单元中获取的人脸识别数据进行 匹配。
步骤 506, 判断是否匹配成功, 如果匹配成功则进入步骤 508 , 否则 进入步骤 510。
步骤 508 , 记录并存储第 N帧画面的时间位置。
步骤 510, 将 N加 100。 即设置预设的抽取步长为 100, 每隔 100帧 图像抽取一个帧图像进行识别。 其中, 预设的抽取步长还可以根据需要自 行设置, 这样不但可以保证识别结果的准确性, 而且可以减少终端的处理 量, 保证处理的速度。
步骤 512 , 判断 N是否大于整个视频的总帧数, 是则进入步骤 514 , 否则进入步骤 504。 N 大于整个视频的总帧数, 则说明这个视频已经识别 完, 此时可以对视频进行处理操作。
步骤 514 , 在播放进度条上显示所有记录的时间位置, 以便于用户进 行进一步的视频处理操作。 (其中, 在记录的相邻的两个时间位置之间的 时间间隔为预设时间间隔时, 两个相邻的时间位置之间的这段间隔也会被 相应的显示出来。 因为预设时间间隔与预设的抽取步长 (即 100 ) 是相互 对应的, 识别出的相邻的帧图像之间的步长为 100 , 则说明相邻的两个帧 图像和中间的 100帧图像都可以与指定图片匹配, 因此, 其时间位置也应 当被显示) 。 这样, 在视频文件的播放界面, 会在时间进度条上突出显示 出为用户查找出的其感兴趣的视频片段, 用户可以直接点击查看到视频文 件中自己感兴趣的部分, 而不需要自己再另外进行查找, 方便了用户的操 作。
另外, 用户还可以将所有自己感兴趣的帧图像合成为一个新的视频文 件, 其合成的顺序可以是按照帧图像的时间顺序, 这样方便用户对新的视 频文件进行查看和存储。
下面结合图 6和图 7详细说明本发明的技术方案。 意图。
如图 6所示, 在视频播放界面, 用户可以点击 "功能开启 602" 按钮 开启图片识别功能。 功能开启之后用户可以在界面上看到 【本地图片】 604 和 【搜索网络图片】 606 两项菜单, 用户可以在其中选择一项, 以进 行指定图片的获取。 此时, 用户可以获取自己想要查看的或感兴趣的人物 的图片, 如用户选择【本地图片】 604 , 并从 【本地图片】 604 中选择图 片 A。 终端获取到图片 A之后, 识别出图片 A 中的特征信息, 并对当前 的视频文件中的帧画面进行识别, 查找出与图片 A相匹配的帧画面。
如图 7所示, 当匹配完毕后, 时间进度条 608上突出显示出用户感兴 趣的角色戏份, 并增加两个控制按钮(前进按钮 704和后退按钮 702 ) 和 一个视频片段合成按钮(片断合成按钮 706 ) 。 用户可以自己选择拖动进 度条或是点击前进按钮 704/后退按钮 702快速定位视频播放自己感兴趣的 角色戏份。 当用户通过点击前进按钮 704 或后退按钮 702 查找角色戏份 时, 终端可以 居当前的播放时间位置, 定位到当前时间位置的前一个或 后一个匹配好的帧画面时间位置处继续播放。 片段合成按钮 706便于用户 可以将感兴趣的视频片段合成一个视频文件并保留。
以上以特定图片为人物图片为例详细说明了本发明的技术方案, 但是 本发明中的特定图片包括但是不限于人物图片, 还可以是其他图像, 如车 牌图像等, 而如果指定图片是车牌图片, 其特征信息就是车牌信息, 则终 端会识别出视频文件中的部分帧图像中的车牌信息进行相应的匹配, 其匹 配及处理过程与上述过程类似, 在此不再赘述。
以上结合附图详细说明了本发明的技术方案, 通过本发明的技术方 案, 可以根据用户的需要在视频文件中自动选择出用户感兴趣的视频画 面, 从而避免用户自己查找的繁瑣操作, 提升用户的使用体验。
根据本发明的实施方式, 还提供了一种存储在非易失性机器可读介质 上的程序产品, 用于视频文件管理, 所述程序产品包括用于使计算机系统 执行以下步骤的机器可执行指令: 对获取到的指定图片进行识别处理, 以 提取出所述指定图片中的特征信息; 根据所述特征信息, 在指定视频文件 中的所有帧图像中按照预设方式查找与所述指定图片相匹配的帧图像, 并 对查找出的所述帧图像进行标记; 根据接收到的处理命令, 对标记的所述 帧图像进行对应的处理操作。
根据本发明的实施方式, 还提供了一种非易失机器可读介质, 存储有 用于视频文件管理的程序产品, 所述程序产品包括用于使计算机系统执行 以下步骤的机器可执行指令: 对获取到的指定图片进行识别处理, 以提取 出所述指定图片中的特征信息; 根据所述特征信息, 在指定视频文件中的 所有帧图像中按照预设方式查找与所述指定图片相匹配的帧图像, 并对查 找出的所述帧图像进行标记; 根据接收到的处理命令, 对标记的所述帧图 像进行对应的处理操作。
根据本发明的实施方式, 还提供了一种机器可读程序, 所述程序使机 器执行如上所述技术方案中任一所述的视频文件管理方法。
根据本发明的实施方式, 还提供了一种存储有机器可读程序的存储介 质, 其中, 所述机器可读程序使得机器执行如上所述技术方案中任一所述 的视频文件管理方法。
以上所述仅为本发明的优选实施例而已, 并不用于限制本发明, 对于 本领域的技术人员来说, 本发明可以有各种更改和变化。 凡在本发明的精 神和原则之内, 所作的任何修改、 等同替换、 改进等, 均应包含在本发明 的保护范围之内。

Claims

权 利 要 求 书
1. 一种终端, 其特征在于, 包括:
图片识别单元, 对获取到的指定图片进行识别处理, 以提取出所述指 定图片中的特征信息;
标记单元, 根据所述特征信息, 在指定视频文件中的所有帧图像中按 照预设方式查找与所述指定图片相匹配的帧图像, 并对查找出的所述帧图 像进行标记;
处理单元, 用于根据接收到的处理命令, 对所述标记单元标记的所述 帧图像进行对应的处理操作。
2. 根据权利要求 1所述的终端, 其特征在于, 所述标记单元包括: 抽取子单元, 用于根据预设的抽取步长在所述指定视频文件的所有帧 图像中抽取出部分帧图像;
识别子单元, 用于对抽取出的所述部分帧图像进行识别, 识别出所述 部分帧图像的图像信息;
比较子单元, 将所述图像信息和所述特征信息进行对比, 并标记出与 所述特征信息相匹配的图像信息所对应的帧图像;
判断子单元, 判断已标记出的帧图像中相邻的两个帧图像之间间隔的 帧图像数是否等于所述预设的抽取步长, 在判断结果为是时, 标记出所述 相邻的两个帧图像之间的帧图像。
3. 根据权利要求 1所述的终端, 其特征在于, 所述处理单元包括: 记录子单元, 记录所述帧图像对应的属性信息, 并根据所述属性信 息, 对所述帧图像进行对应的处理操作, 其中, 所述属性信息中包含时间 信息和内容信息。
4. 根据权利要求 3 所述的终端, 其特征在于, 所述记录子单元还用 于:
根据所述时间信息, 确定所述帧图像在所述指定视频文件对应的时间 进度条上的时间点; 以及
所述处理单元还包括: 显示子单元, 用于显示所述时间进度条, 并在所述时间进度条上突出 显示出所述时间点。
5. 根据权利要求 4 所述的终端, 其特征在于, 所述显示子单元还用 于:
根据接收到的选择指令, 在被突出显示的所有时间点中选择一时间 点, 并确定与选择的一时间点相邻的前一时间点和 /或后一时间点, 按照 时间顺序显示所选择的一时间点和所述前一时间点之间或所选择的一时间 点和所述后一时间点之间所对应的帧图像。
6. 根据权利要求 3 所述的终端, 其特征在于, 所述处理单元还包 括:
视频生成子单元, 根据接收到的视频生成指令, 获取所述标记单元标 记的所有的帧图像, 并按照所述所有的帧图像中每个帧图像对应的属性信 息, 将所述所有的帧图像重新组合成一个新的视频文件; 以及
存储子单元, 用于存储所述新的视频文件。
7. 根据权利要求 1 所述的终端, 其特征在于, 所述图片识别单元还 用于:
根据接收到的图片获取指令, 从本地和 /或互联网上获取所述指定图 片。
8. 根据权利要求 1 至 7 中任一项所述的终端, 其特征在于, 还包 括:
控制单元, 用于在接收到功能开启或关闭命令时, 控制所述图片识别 单元开始或停止对获取到的所述指定图片进行识别处理。
9. 一种视频文件管理方法, 其特征在于, 包括:
步骤 202 , 对获取到的指定图片进行识别处理, 以提取出所述指定图 片中的特征信息;
步骤 204 , 根据所述特征信息, 在指定视频文件中的所有帧图像中按 照预设方式查找与所述指定图片相匹配的帧图像, 并对查找出的所述帧图 像进行标记;
步骤 206 , 根据接收到的处理命令, 对标记的所述帧图像进行对应的 处理操作。
10. 根据权利要求 9所述的视频文件管理方法, 其特征在于, 所述步 骤 204包括:
根据预设的抽取步长在所述指定视频文件的所有帧图像中抽取出部分 帧图像;
对抽取出的所述部分帧图像进行识别, 识别出所述部分帧图像的图像 信息; 以及
将所述图像信息和所述特征信息进行对比, 并标记出与所述特征信息 相匹配的图像信息所对应的帧图像;
判断已标记出的帧图像中相邻的两个帧图像之间间隔的帧图像数是否 等于所述预设的抽取步长, 在判断结果为是时, 标记出所述相邻的两个帧 图像之间的帧图像。
11. 根据权利要求 9所述的视频文件管理方法, 其特征在于, 所述步 骤 206包括:
记录所述帧图像对应的属性信息, 并根据所述属性信息, 对所述帧图 像进行对应的处理操作, 其中, 所述属性信息中包含时间信息和内容信 息。
12. 根据权利要求 11 所述的视频文件管理方法, 其特征在于, 所述 步骤 206还包括:
根据所述时间信息, 确定所述帧图像在所述指定视频文件对应的时间 进度条上的时间点; 以及
显示所述时间进度条, 并在所述时间进度条上突出显示出所述时间 点。
13. 根据权利要求 12 所述的视频文件管理方法, 其特征在于, 所述 步骤 206还包括:
根据接收到的选择指令, 在被突出显示的所有时间点中选择一时间 点, 并确定与选择的一时间点相邻的前一时间点和 /或后一时间点, 按照 时间顺序显示所选择的一时间点和所述前一时间点之间或所选择的一时间 点和所述后一时间点之间所对应的帧图像。
14. 根据权利要求 11 所述的视频文件管理方法, 其特征在于, 所述 步骤 206还包括:
根据接收到的视频生成指令, 获取标记的所有的帧图像, 并按照所述 所有的帧图像中每个帧图像对应的属性信息, 将所述所有的帧图像重新组 合成一个新的视频文件; 以及
存储所述新的视频文件。
15. 根据权利要求 9 所述的视频文件管理方法, 其特征在于, 还包 括:
根据接收到的图片获取指令, 从本地和 /或互联网上获取所述指定图 片。
16. 根据权利要求 9 至 15 中任一项所述的视频文件管理方法, 其特 征在于, 还包括:
在接收到功能开启或关闭命令时, 开始或停止对获取到的所述指定图 片进行识别处理。
PCT/CN2013/086207 2013-10-30 2013-10-30 终端和视频文件管理方法 WO2015061979A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP13896763.3A EP3065079A4 (en) 2013-10-30 2013-10-30 Terminal and method for managing video file
PCT/CN2013/086207 WO2015061979A1 (zh) 2013-10-30 2013-10-30 终端和视频文件管理方法
CN201380068861.7A CN104995639B (zh) 2013-10-30 2013-10-30 终端和视频文件管理方法
US15/025,775 US10229323B2 (en) 2013-10-30 2013-10-30 Terminal and method for managing video file

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2013/086207 WO2015061979A1 (zh) 2013-10-30 2013-10-30 终端和视频文件管理方法

Publications (1)

Publication Number Publication Date
WO2015061979A1 true WO2015061979A1 (zh) 2015-05-07

Family

ID=53003110

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/086207 WO2015061979A1 (zh) 2013-10-30 2013-10-30 终端和视频文件管理方法

Country Status (4)

Country Link
US (1) US10229323B2 (zh)
EP (1) EP3065079A4 (zh)
CN (1) CN104995639B (zh)
WO (1) WO2015061979A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106851407A (zh) * 2017-01-24 2017-06-13 维沃移动通信有限公司 一种视频播放进度的控制方法及终端
CN108810617A (zh) * 2018-06-12 2018-11-13 优视科技有限公司 一种根据视频制作图像海报的方法、装置和终端设备

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104038848A (zh) * 2014-05-30 2014-09-10 无锡天脉聚源传媒科技有限公司 一种视频处理方法及装置
CN106991362A (zh) * 2016-01-21 2017-07-28 腾讯科技(深圳)有限公司 一种针对视频的图片处理方法及装置
CN110113677A (zh) * 2018-02-01 2019-08-09 阿里巴巴集团控股有限公司 视频主题的生成方法和装置
CN108769801B (zh) * 2018-05-28 2019-03-29 广州虎牙信息科技有限公司 短视频的合成方法、装置、设备及存储介质
CN110472488A (zh) * 2019-07-03 2019-11-19 平安科技(深圳)有限公司 基于数据处理的图片显示方法、装置和计算机设备
CN110569392B (zh) * 2019-08-28 2023-01-10 深圳市天视通技术有限公司 一种多视频处理系统和方法
CN110580508A (zh) * 2019-09-06 2019-12-17 捷开通讯(深圳)有限公司 视频分类方法、装置、存储介质和移动终端
CN111050214A (zh) * 2019-12-26 2020-04-21 维沃移动通信有限公司 一种视频播放方法及电子设备
CN111400552B (zh) * 2020-03-31 2024-02-27 维沃移动通信有限公司 便签创建方法及电子设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090103887A1 (en) * 2007-10-22 2009-04-23 Samsung Electronics Co., Ltd. Video tagging method and video apparatus using the same
CN101930779A (zh) * 2010-07-29 2010-12-29 华为终端有限公司 一种视频批注方法及视频播放器
CN102222103A (zh) * 2011-06-22 2011-10-19 央视国际网络有限公司 视频内容的匹配关系的处理方法及装置
CN102549603A (zh) * 2009-08-24 2012-07-04 谷歌公司 基于相关性的图像选择

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040128317A1 (en) * 2000-07-24 2004-07-01 Sanghoon Sull Methods and apparatuses for viewing, browsing, navigating and bookmarking videos and displaying images
KR101437653B1 (ko) * 2007-01-26 2014-09-12 삼성전자주식회사 재생시점 변경용 gui 제공방법 및 이를 적용한 영상기기
CN101425082A (zh) * 2008-12-15 2009-05-06 深圳市迅雷网络技术有限公司 视频文件内容确定方法及系统
CN101448100B (zh) * 2008-12-26 2011-04-06 西安交通大学 一种快速准确的视频字幕提取方法
CN102930553B (zh) * 2011-08-10 2016-03-30 中国移动通信集团上海有限公司 不良视频内容识别方法及装置
KR101341808B1 (ko) * 2011-11-30 2013-12-17 고려대학교 산학협력단 영상 내 비주얼 특징을 이용한 영상 요약 방법 및 시스템
CN102611685B (zh) * 2011-12-16 2018-08-14 中兴通讯股份有限公司 一种标记流媒体的方法、装置及系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090103887A1 (en) * 2007-10-22 2009-04-23 Samsung Electronics Co., Ltd. Video tagging method and video apparatus using the same
CN102549603A (zh) * 2009-08-24 2012-07-04 谷歌公司 基于相关性的图像选择
CN101930779A (zh) * 2010-07-29 2010-12-29 华为终端有限公司 一种视频批注方法及视频播放器
CN102222103A (zh) * 2011-06-22 2011-10-19 央视国际网络有限公司 视频内容的匹配关系的处理方法及装置

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106851407A (zh) * 2017-01-24 2017-06-13 维沃移动通信有限公司 一种视频播放进度的控制方法及终端
CN108810617A (zh) * 2018-06-12 2018-11-13 优视科技有限公司 一种根据视频制作图像海报的方法、装置和终端设备
WO2019237966A1 (zh) * 2018-06-12 2019-12-19 阿里巴巴集团控股有限公司 一种根据视频制作图像海报的方法、装置和终端设备

Also Published As

Publication number Publication date
EP3065079A1 (en) 2016-09-07
US10229323B2 (en) 2019-03-12
CN104995639B (zh) 2018-11-06
US20160247025A1 (en) 2016-08-25
CN104995639A (zh) 2015-10-21
EP3065079A4 (en) 2017-06-21

Similar Documents

Publication Publication Date Title
WO2015061979A1 (zh) 终端和视频文件管理方法
JP5791364B2 (ja) 顔認識装置、顔認識方法、顔認識プログラム、およびそのプログラムを記録した記録媒体
JP2020536455A5 (zh)
US7908556B2 (en) Method and system for media landmark identification
US9881215B2 (en) Apparatus and method for identifying a still image contained in moving image contents
JP2007281858A (ja) 動画編集装置
JP2006054746A5 (zh)
US20100121875A1 (en) Information processing apparatus, information processing method, and information processing program
WO2019129075A1 (zh) 视频检索的方法和装置以及计算机可读存储介质
WO2016165346A1 (zh) 存储和播放音频文件的方法和装置
CN113411516A (zh) 视频处理方法、装置、电子设备及存储介质
JP2010086221A (ja) 画像編集方法および装置、ならびに画像編集方法を実行させるプログラムを記録したコンピュータに読取可能な記録媒体
JP2008109289A5 (zh)
Pongnumkul et al. Creating map-based storyboards for browsing tour videos
WO2017008498A1 (zh) 搜索节目的方法及装置
JP6276570B2 (ja) 画像音声再生システム及び画像音声再生方法とプログラム
WO2019015411A1 (zh) 录屏方法、装置及电子设备
JP2007124368A (ja) セグメントメタデータ作成装置およびセグメントメタデータ作成方法
JP2009283020A (ja) 記録装置、再生装置、及びプログラム
JP2009081733A (ja) 画像再生装置、及びプログラム
JP2004297305A (ja) データベース構築装置、データベース構築プログラム、画像検索装置、画像検索プログラム、及び画像記録再生装置
CN113573096A (zh) 视频处理方法、装置、电子设备及介质
JP6256738B2 (ja) 動画選択装置、動画選択方法とプログラム
JP6261198B2 (ja) 情報処理装置、情報処理方法およびプログラム
CN113747233A (zh) 一种音乐替换方法、装置、电子设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13896763

Country of ref document: EP

Kind code of ref document: A1

REEP Request for entry into the european phase

Ref document number: 2013896763

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2013896763

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 15025775

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE