WO2015117572A1 - Labelling method for moving objects of concentrated video, and playing method and device - Google Patents

Labelling method for moving objects of concentrated video, and playing method and device Download PDF

Info

Publication number
WO2015117572A1
WO2015117572A1 PCT/CN2015/072793 CN2015072793W WO2015117572A1 WO 2015117572 A1 WO2015117572 A1 WO 2015117572A1 CN 2015072793 W CN2015072793 W CN 2015072793W WO 2015117572 A1 WO2015117572 A1 WO 2015117572A1
Authority
WO
WIPO (PCT)
Prior art keywords
video frame
moving
concentrated
concentrated video
moving target
Prior art date
Application number
PCT/CN2015/072793
Other languages
French (fr)
Chinese (zh)
Inventor
李辉
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2015117572A1 publication Critical patent/WO2015117572A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
    • H04N21/84—Generation or processing of descriptive data, e.g. content descriptors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/236—Assembling of a multiplex stream, e.g. transport stream, by combining a video stream with other content or additional data, e.g. inserting a URL [Uniform Resource Locator] into a video stream, multiplexing software data into a video stream; Remultiplexing of multiplex streams; Insertion of stuffing bits into the multiplex stream, e.g. to obtain a constant bit-rate; Assembling of a packetised elementary stream
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/81—Monomedia components thereof
    • H04N21/8146—Monomedia components thereof involving graphical data, e.g. 3D object, 2D graphics

Definitions

  • Embodiments of the present invention relate to the field of video processing technologies, and in particular, to a moving target labeling method, a playing method, and a device for a concentrated video.
  • Video enrichment not only concentrates on the essence of the event, but also the activity event. The video with no value will be eliminated.
  • One is to synthesize the video frame data and the moving target annotation of the video frame during the video concentration analysis process.
  • the disadvantage of this processing method is that the processing performance is expensive, and the annotation information cannot be separated from the concentrated video frame.
  • the other is that in the process of video concentration analysis, the annotation information of the moving target of the video frame is written in the description file; when the player plays the concentrated video, the moving target of the video frame is marked according to the information of the description file.
  • the disadvantage of this processing method is that the description file is relatively large, which is not conducive to saving and transmitting. Moreover, during playback, the moving target of each frame of video is marked, and information search is needed in the description file, and then the player finds according to the search. Information is annotated and affects playback performance.
  • an embodiment of the present invention provides a moving target labeling method, a playing method, and a device for concentrating video, which can separate the concentrated video frame data and the labeling information, and the labeling information can be conveniently saved and transmitted with the video information. Thereby improving the labeling efficiency and ensuring the playback performance.
  • the present invention provides a moving target labeling method for a concentrated video, comprising: acquiring annotation information of all moving targets in a concentrated video frame and relative playing time of the concentrated video frame;
  • Encapsulating the enriched video frame and the labeling information of all moving targets of the concentrated video frame respectively Media packets and annotated packets;
  • it also includes:
  • the media data package, the labeling information packet, and the association between the media data packet and the labeling information package are respectively saved in the concentrated video file to obtain a target concentrated video file.
  • the acquiring the annotation information of all the moving targets in a concentrated video frame and the relative playing time of the concentrated video frames include:
  • the annotation information of the moving target includes: coordinates of the moving target, a height of the moving target, and a width of the moving target.
  • the encapsulating the condensed video frame into a media data packet comprises: packaging the condensed video frame into a media data packet in a first preset format.
  • the encapsulating the annotation information of all moving targets of the concentrated video frame into the annotation information package includes:
  • the invention also provides a method for playing a concentrated video, comprising:
  • the annotation information of the moving target includes: coordinates of the moving target, height of the moving target, and width of the moving target.
  • the invention further provides a moving target marking device for concentrating video, comprising:
  • An obtaining module configured to obtain annotation information of all moving targets in a concentrated video frame and a relative playing time of the concentrated video frame
  • the encapsulation module is configured to encapsulate the enrichment video frame and the annotation information of all the moving targets of the enriched video frame into a media data packet and an annotation information packet, respectively;
  • the association module is configured to establish an association between the media data packet of the concentrated video frame and the labeling information packet according to the relative playing time of the concentrated video frame.
  • it also includes:
  • the saving module is configured to save the media data package, the labeling information packet, and the association between the media data packet and the labeling information package to the concentrated video file to obtain the target concentrated video file.
  • the obtaining module includes:
  • the concentrating module is configured to perform concentration processing on the original video file to obtain a concentrated video frame
  • An extraction module configured to perform a moving target analysis on the concentrated video frame, and extract a moving target; wherein the moving target includes: a moving sub-target that does not overlap with other moving sub-objects in the concentrated video frame, and an overlapping relationship Multiple sports sub-goals;
  • the determining module is configured to acquire the labeling information of each moving target separately, and determine the labeling information of all the moving targets.
  • the encapsulating module at least includes:
  • the first encapsulation submodule is configured to encapsulate the enriched video frame into a media packet of a first preset format.
  • the encapsulating module further includes:
  • the second encapsulation submodule is configured to encapsulate the annotation information of all the moving targets of the concentrated video frame into the annotation information packet of the second preset format.
  • the present invention further provides a playback device for a concentrated video, comprising:
  • a first parsing module configured to parse the target concentrated video file to obtain a concentrated video frame and Relative play time of the concentrated video frame
  • a first acquiring module configured to acquire, according to the relative playing time, an annotation information packet associated with all moving targets of the concentrated video frame
  • a second parsing module configured to parse the label information packet, and determine labeling information of a moving target of the concentrated video frame
  • the playing module is configured to superimpose and display the enriched video frame and the labeling information of the moving target of the concentrated video frame according to the relative playing time to complete the playing.
  • the annotation information of the moving target includes: coordinates of the moving target, a height of the moving target, and a width of the moving target.
  • the labeling information of the moving target of the condensed video frame and the condensed video frame data are separately encapsulated and written into the condensed video file, thereby reducing the step of synthesizing the two.
  • the processing efficiency is improved, and the separation of the video frame data and the annotation information is realized.
  • the labeling information of the video frame and the moving target of the video is associated by the relative playing time of the video frame, so that the labeling information is convenient.
  • the storage and transmission of the video information directly realizes the labeling of the moving target in the process of concentrating the video playing, thereby improving the labeling efficiency and ensuring the playing efficiency.
  • FIG. 1 is a flowchart of a method for marking a moving target of a concentrated video according to an embodiment of the present invention
  • FIG. 2 is a flowchart of determining an association relationship of a moving target labeling method for a concentrated video according to an embodiment of the present invention
  • FIG. 3 is a flowchart of a method for playing a concentrated video in an embodiment of the present invention
  • FIG. 4 is a schematic diagram of a specific labeling and playing process of an MP4 file as an example in the embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a moving target marking device for concentrating video according to an embodiment of the present invention
  • FIG. 6 is a schematic structural diagram of a device for playing back a concentrated video according to an embodiment of the present invention.
  • the present invention is directed to the prior art that the concentrated video technology has high processing performance consumption, and the annotation information cannot be separated from the concentrated video frame or the moving object of the video frame is processed by the description file, but the description file is large, which is not conducive to preservation and transmission.
  • the invention provides a moving target labeling method, a playing method and a device for concentrating video.
  • the labeling information of the moving target of the concentrated video frame and the concentrated video frame data are separately encapsulated and written into the concentrated video file, thereby reducing
  • the step of synthesizing the two improves the processing efficiency, and realizes the separation of the video frame data and the annotation information.
  • the embodiment of the present invention associates the video frame with the annotation information of the moving target of the video by the relative playing time of the video frame.
  • the annotation information is conveniently saved and transmitted with the video information, and the moving target is directly marked in the concentrated video playback process, thereby improving the labeling efficiency and ensuring the playback efficiency.
  • an embodiment of the present invention provides a method for marking a moving target of a concentrated video, including:
  • Step 100 Obtain annotation information of all moving targets in a concentrated video frame and relative playing time of the concentrated video frame.
  • Step 101 Encapsulate the obtained concentrated video frame and the labeling information of all moving targets of the concentrated video frame into a media data packet and an annotation information packet, respectively;
  • Step 102 Establish an association between the media data packet of the concentrated video frame and the labeled information packet according to the relative playing time of the obtained concentrated video frame.
  • the moving target in step 100 refers to a moving object in a video frame, such as a moving person, a vehicle, etc.; an arrow, a circle, or the like for the moving target can help the person watching the video in a short time. Read all the activity goals in time.
  • the method for labeling the moving target of the condensed video in the embodiment of the present invention is performed in the process of concentrating the original video file, that is, the original video file is condensed to concentrate one frame of the video frame, and then the following steps are performed. 101, step 102, until the concentration of the original video file is completed, the extraction of the annotation information of the moving target is also completed, and the annotation is directly called when the concentrated video is played.
  • the information directly marks the moving target, and improves the labeling efficiency; at the same time, the concentrated video frame obtained in step 101 and the labeling information of the moving target of the concentrated video are separately packaged, respectively, and the concentrated video file is written to make the video frame data and
  • the labeling information achieves separation, saving two combined processing steps and improving processing efficiency.
  • the labeling information of the moving video frame and the moving target of the concentrated video frame are respectively encapsulated, it is necessary to establish an association between the concentrated video frame and the labeling information of the moving target of the concentrated video frame according to the relative playing time, which is convenient.
  • the video is played, the corresponding concentrated video frame and the annotation information of the moving target of the concentrated video are found through the association relationship, and are combined and played to form a complete video.
  • the moving target labeling method of the concentrated video further includes:
  • the established media data package, the labeled information package, and the association between the media data package and the labeled information package are respectively saved into the concentrated video file to obtain the target concentrated video file.
  • the media data packet and the annotation information packet of the next concentrated video frame are continuously obtained. And the association between the media packet and the annotated packet until all of the condensed video frames have been processed, and the media packets, annotated packets, and the association between the media packets and the annotated packets for each frame Save separately to get the target concentrated video file.
  • the condensed video file of the purpose is a file obtained by labeling a moving target by using the moving target labeling method provided by the embodiment of the present invention.
  • association relationship is written into the metadata description part of the condensed video file, and the metadata description part is generally used to store the association between the various parts in the video file; for example, the normal video file generally includes at least an audio track and a video.
  • the metadata description section In order to ensure that audio and video are played synchronously during playback, it is necessary to store the relationship between audio and video in the metadata description section, that is, which corresponding audio file should be played simultaneously when playing the video.
  • step 100 specifically includes:
  • Step 200 Perform concentration processing on the original video file to obtain a concentrated video frame.
  • Step 201 Perform moving object analysis on the obtained concentrated video frame to extract a moving target, where the moving target includes: a moving sub-target that does not overlap with other moving sub-objects in the concentrated video frame, and a plurality of sports sub-objects with overlapping relationship aims;
  • Step 202 Acquire label information of each moving target separately, and determine labeling information of all moving targets.
  • the concentrating process in step 200 is specifically: the specific process of concentrating the preset number of frame images may be performed by a concentrating algorithm, for example, if the preset number is 5 frames, the input will be input.
  • the image of the original video file of the 5 frames is processed by a condensing algorithm, and a video image of the frame is output, and the video image of the frame is a condensed video frame image of the embodiment of the present invention, that is, the condensed video frame is obtained; It is the essence of the above 5 frames of the original video file image, and the condensed video frame is obtained by merging the valuable video and combining the valuable video.
  • the preset number can be set according to different needs of the user. For example, if the user needs to condense one video into 10M and the user needs to condense the same video into 5M, a different preset number can be selected; of course, the smaller the concentrated video is. , the larger the value of the preset number.
  • each moving object appearing in the concentrated video frame may be regarded as a moving sub-target, and then the moving sub-target that does not overlap with other moving sub-objects is determined as the moving target.
  • a plurality of sports sub-goals having overlapping relationship with other sports sub-goals are collectively determined as one moving target. For example, in a concentrated video frame, there are two moving sub-targets, a moving sub-target A and a moving sub-target B. If the moving sub-target A and the moving sub-target B do not overlap in the concentrated video frame, the moving sub-target A is determined.
  • step 202 the same or different annotation information is determined for each moving target, and saved for subsequent calls.
  • the labeling information of the moving target includes: a coordinate of the moving target, a height of the moving target, and a width of the moving target.
  • the coordinates of the moving target (X-axis, Y-axis, and the Z-axis of the stereo), the height of the moving target, and the width of the moving target need to be acquired.
  • step 101 further includes: encapsulating the concentrated video frame into a first Media packets in a preset format.
  • the media data packet includes the concentrated video frame
  • the encapsulation process is mainly to convert the frame data into a frame format conforming to the playback format of the concentrated video file, such as mp4 format, rmvb format, mtv format, wmv format, etc. Etc., determine the format of its media packet based on the format of the condensed video file.
  • the step 102 further includes: encapsulating the annotation information of all the moving targets of the concentrated video frame into the annotation information packets of the second preset format. among them,
  • the labeling information packet includes the labeling information of all the moving objects in the concentrated video frame, and the packaging process mainly packs the labeling information into a file conforming to the labeling information format of the concentrated video file, such as the labeling information track of the playing file in the mp4 format.
  • the format, the format of the labeling information track of the playback file in the rmvb format, and the like are not listed here.
  • the media data packet and the labeled information packet of the frame are played when the preset time point is 5 s, the media data packet and the labeled information at the relative playing time are determined. The association of the package.
  • the moving target labeling method of the present invention is marked in the enrichment process of the original video file, when each concentrated video frame is processed, the concentrated video file at this time is the target concentrated video file. And the concentration and corresponding label information are saved, which improves the concentration efficiency.
  • an embodiment of the present invention further provides a method for playing a concentrated video, including:
  • Step 300 parsing the target concentrated video file to obtain a concentrated video frame and a relative playing time of the concentrated video frame
  • Step 301 Acquire an annotation information packet associated with all moving targets of the concentrated video frame according to the relative playing time
  • Step 302 parsing the label information packet, and determining labeling information of a moving target of the concentrated video frame
  • Step 303 Perform superimposed display processing on the labeled information of the concentrated video frame and the moving target of the concentrated video frame according to the relative playing time, and complete the playing.
  • the playing process of the concentrated video is the concentrated video provided by the present invention.
  • the method of labeling the moving object corresponds to; the essence is opposite to the process of the labeling method of the moving object of the concentrated video provided by the present invention.
  • the annotation information of the moving target includes: coordinates of the moving target, a height of the moving target, and a width of the moving target.
  • the original video file is an MP4 file
  • the moving target labeling method and the corresponding concentrated video playing process provided by the present invention are specifically described:
  • Steps for generating the moving target annotation information in the MP4 concentrated video are identical to Steps for generating the moving target annotation information in the MP4 concentrated video:
  • Step 401 Perform MP4 analysis processing on the MP4 concentrated video file, parse out one frame of concentrated video frame data, and perform a concentration algorithm processing;
  • Step 402 Concentration algorithm processing, if there is no output data, returning to step 401, continuing to parse the file; if there is data output, outputting the data;
  • Step 403 The video frame data after the concentration processing
  • Step 404 labeling information such as an X-axis, a Y-axis, a Heigh height, and a Width width of the moving target of the video frame after the concentration processing;
  • Step 405 Encapsulate step 403 and step 404 into a video track and an information track in the MP4 file respectively, and package according to the relative play time, so as to correspondingly associate in the playing;
  • Step 406 Write the encapsulated video track and the information track to the MP4 concentrated video file respectively, and simultaneously write the information related to the video track and the information track with respect to the play time to the MP4 file metadata description part. Returning to step 401 processing until the original MP4 file is parsed.
  • the playing steps of the moving target annotation information in the MP4 concentrated video include:
  • Step 407 Parsing the MP4 concentrated video file, parsing the video frame data and the relative playing time of the frame, and finding the corresponding flag information according to the relative playing time;
  • Step 408 Obtain video frame data and relative play time.
  • Step 409 Obtain information information of the X-axis, Y-axis, Heigh, and Width of the information track and relative play time;
  • Step 410 Associate the video frame data with the labeling information of the moving target of the frame according to the relative playing time
  • Step 411 When the player plays, the video frame data and the moving target label of the frame are superimposed; the looping to step 407 is performed until the MP4 concentrated video file is parsed, and the playing is completed.
  • Video moving target marking device including:
  • the obtaining module 10 is configured to obtain annotation information of all moving targets in a concentrated video frame and a relative playing time of the concentrated video frame;
  • the encapsulation module 20 is configured to encapsulate the enrichment video frame and the annotation information of all moving targets of the enriched video frame into a media data packet and an annotation information packet, respectively;
  • the association module 30 is configured to establish an association between the media data packet of the concentrated video frame and the annotation information packet according to the relative play time of the concentrated video frame.
  • the apparatus further includes a saving module (not shown in FIG. 5) configured to associate the media data packet, the labeling information packet, and the media data packet with the labeling information packet, Save them separately to the concentrated video file to get the target concentrated video file.
  • a saving module (not shown in FIG. 5) configured to associate the media data packet, the labeling information packet, and the media data packet with the labeling information packet, Save them separately to the concentrated video file to get the target concentrated video file.
  • the acquiring module 10 may specifically include:
  • the concentrating module is configured to perform concentration processing on the original video file to obtain a concentrated video frame
  • An extraction module configured to perform a moving target analysis on the concentrated video frame, and extract a moving target; wherein the moving target includes: a moving sub-target that does not overlap with other moving sub-objects in the concentrated video frame, and an overlapping relationship Multiple sports sub-goals;
  • the determining module is configured to acquire the labeling information of each moving target separately, and determine the labeling information of all the moving targets.
  • the encapsulating module 20 may specifically include:
  • the first encapsulation submodule is configured to encapsulate the enriched video frame into a media packet of a first preset format.
  • the package module 20 further includes:
  • the second encapsulation submodule is configured to encapsulate the annotation information of all the moving targets of the concentrated video frame into the annotation information packet of the second preset format.
  • the labeling information of the moving target of the condensed video frame and the condensed video frame data are separately encapsulated and written into the condensed video file, thereby reducing the steps of synthesizing the two.
  • the processing efficiency is improved, and the separation of the video frame data and the annotation information is realized.
  • the labeling information of the video frame and the moving target of the video is associated by the relative playing time of the video frame, so that the labeling information is convenient.
  • video information storage and transmission, in the process of concentrated video playback directly achieve the labeling of moving targets, thereby improving The labeling efficiency ensures the playback efficiency.
  • the embodiment of the present invention further provides a playback device for a concentrated video, including:
  • the first parsing module 60 is configured to parse the target concentrated video file to obtain a concentrated video frame and a relative playing time of the concentrated video frame on a preset time axis;
  • the first obtaining module 70 is configured to acquire, according to the relative playing time, an annotated information packet associated with all moving targets of the concentrated video frame;
  • a second parsing module 80 configured to parse the label information packet, and determine label information of a moving target of the concentrated video frame
  • the playing module 90 is configured to perform superimposed display processing on the labeling information of the moving video frame and the moving target of the concentrated video frame according to the relative playing time to complete the playing.
  • the annotation information of the moving target includes: coordinates of the moving target, height of the moving target, and width of the moving target.
  • the annotation information of all the moving objects in a concentrated video frame and the relative playing time of the concentrated video frame are obtained in the embodiment of the present invention; Encapsulating information of the moving video frame and all moving targets of the concentrated video frame into a media data packet and an annotation information packet respectively; establishing a media data packet of the concentrated video frame according to a relative playing time of the concentrated video frame And the association between the tagged packets.
  • the embodiment of the present invention associates the video frame with the labeling information of the moving target of the video by the relative playing time of the video frame, so that the labeling information is conveniently saved and transmitted with the video information, and is directly in the concentrated video playing process.
  • the labeling of the moving target is realized, thereby improving the labeling efficiency and ensuring the playing efficiency.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Computer Graphics (AREA)
  • Television Signal Processing For Recording (AREA)

Abstract

A labelling method for moving objects of a concentrated video, and a playing method and device. The method in the embodiments of the present invention comprises: acquiring labelling information about all moving objects in a concentrated video frame and relative playing time of the concentrated video frame; respectively encapsulating the concentrated video frame and the labelling information about all the moving objects in the concentrated video frame into a media data packet and a labelling information packet; and according to the relative playing time of the concentrated video frame, establishing a correlation between the media data packet and the labelling information packet of the concentrated video frame. By respectively encapsulating and writing labelling information about moving objects of a concentrated video frame and data of the concentrated video frame into a concentrated video file, the step of synthesizing the two is omitted, the processing efficiency is improved, and the labelling information can be conveniently saved and transmitted together with the video information, such that the moving objects are directly labelled in the process of playing the concentrated video, thereby improving the labelling efficiency and guaranteeing the playing efficiency.

Description

一种浓缩视频的运动目标标注方法、播放方法及装置Moving target marking method, playing method and device for concentrated video 技术领域Technical field
本发明实施例涉及视频处理技术领域，特别涉及一种浓缩视频的运动目标标注方法、播放方法及装置。Embodiments of the present invention relate to the field of video processing technologies, and in particular, to a moving target labeling method, a playing method, and a device for a concentrated video.
背景技术Background technique
视频浓缩不仅浓缩的是事件的精华，也是活动事件的全部，没有价值的视频将被剔除；通过分析合并技术，可以在很短的时间中看完所有的活动目标成为可能，且对运动目标进行箭头，圆圈等标注。Video enrichment not only concentrates on the essence of the event, but also the activity event. The video with no value will be eliminated. By analyzing the merge technology, it is possible to see all the activity targets in a short time, and to carry out the moving target. Arrows, circles, etc.
目前，常有对浓缩视频中运动目标标注的方法有两种：At present, there are two methods for labeling moving targets in concentrated video:
一种是在视频浓缩分析过程中，对视频帧数据和该视频帧的运动目标标注进行合成处理。这种处理方法的缺点是：处理性能消耗大，标注信息与浓缩视频帧不能分离。One is to synthesize the video frame data and the moving target annotation of the video frame during the video concentration analysis process. The disadvantage of this processing method is that the processing performance is expensive, and the annotation information cannot be separated from the concentrated video frame.
另一种是在视频浓缩分析过程中，把视频帧的运动目标的标注信息写在描述文件中；播放器播放浓缩视频时，根据描述文件的信息对视频帧的运动目标进行标注处理。这种处理方法的缺点是：描述文件比较大，不利于保存和传输；而且在播放时，对每帧视频的运动目标标注，需要在描述文件中进行信息查找，然后播放器再根据查找到的信息进行标注处理，影响了播放性能。The other is that in the process of video concentration analysis, the annotation information of the moving target of the video frame is written in the description file; when the player plays the concentrated video, the moving target of the video frame is marked according to the information of the description file. The disadvantage of this processing method is that the description file is relatively large, which is not conducive to saving and transmitting. Moreover, during playback, the moving target of each frame of video is marked, and information search is needed in the description file, and then the player finds according to the search. Information is annotated and affects playback performance.
发明内容Summary of the invention
为了解决上述技术问题，本发明实施例提供一种浓缩视频的运动目标标注方法、播放方法及装置，能够使浓缩视频帧数据和标注信息分离，且标注信息可以方便地与视频信息保存与传送，从而提高标注效率，保证播放性能。In order to solve the above technical problem, an embodiment of the present invention provides a moving target labeling method, a playing method, and a device for concentrating video, which can separate the concentrated video frame data and the labeling information, and the labeling information can be conveniently saved and transmitted with the video information. Thereby improving the labeling efficiency and ensuring the playback performance.
为了达到上述技术目的，本发明提供一种浓缩视频的运动目标标注方法，包括：获取一浓缩视频帧中的所有运动目标的标注信息以及该浓缩视频帧的相对播放时间；In order to achieve the above technical purpose, the present invention provides a moving target labeling method for a concentrated video, comprising: acquiring annotation information of all moving targets in a concentrated video frame and relative playing time of the concentrated video frame;
将所述浓缩视频帧和该浓缩视频帧的所有运动目标的标注信息分别封装 成媒体数据包和标注信息包；Encapsulating the enriched video frame and the labeling information of all moving targets of the concentrated video frame respectively Media packets and annotated packets;
根据所述浓缩视频帧的相对播放时间，建立所述浓缩视频帧的媒体数据包和标注信息包之间的关联。Establishing an association between the media data packet of the condensed video frame and the tagged information packet according to the relative play time of the condensed video frame.
可选地，还包括：Optionally, it also includes:
将所述媒体数据包、标注信息包，以及所述媒体数据包与标注信息包之间的关联，分别保存到浓缩视频文件中，得到目的浓缩视频文件。The media data package, the labeling information packet, and the association between the media data packet and the labeling information package are respectively saved in the concentrated video file to obtain a target concentrated video file.
可选地，所述获取一浓缩视频帧中的所有运动目标的标注信息以及该浓缩视频帧的相对播放时间包括：Optionally, the acquiring the annotation information of all the moving targets in a concentrated video frame and the relative playing time of the concentrated video frames include:
对原视频文件进行浓缩处理，得到一所述浓缩视频帧；Concentrating the original video file to obtain a concentrated video frame;
对所述浓缩视频帧进行运动目标分析，提取运动目标；其中，运动目标包括：在所述浓缩视频帧中没有与其他运动子目标重叠的运动子目标，以及具有重叠关系的多个运动子目标；Performing a moving target analysis on the concentrated video frame to extract a moving target; wherein the moving target includes: a moving sub-target overlapping with other moving sub-objects in the concentrated video frame, and a plurality of moving sub-objects having overlapping relationships ;
分别获取每一个运动目标的标注信息，确定所述所有运动目标的标注信息。Obtaining the annotation information of each moving target separately, and determining the labeling information of all the moving targets.
可选地，所述运动目标的标注信息包括：运动目标的坐标、运动目标的高度和运动目标的宽度。Optionally, the annotation information of the moving target includes: coordinates of the moving target, a height of the moving target, and a width of the moving target.
可选地，所述将所述浓缩视频帧封装成媒体数据包包括：将所述浓缩视频帧封装成第一预设格式的媒体数据包。Optionally, the encapsulating the condensed video frame into a media data packet comprises: packaging the condensed video frame into a media data packet in a first preset format.
可选地，所述将该浓缩视频帧的所有运动目标的标注信息封装成标注信息包包括：Optionally, the encapsulating the annotation information of all moving targets of the concentrated video frame into the annotation information package includes:
将所述浓缩视频帧的所有运动目标的标注信息封装成第二预设格式的标注信息包。And labeling information of all moving targets of the concentrated video frame into an annotation information packet of a second preset format.
本发明还提供了一种浓缩视频的播放方法，包括：The invention also provides a method for playing a concentrated video, comprising:
解析目的浓缩视频文件，得到一浓缩视频帧以及该浓缩视频帧的相对播放时间；Parsing the target concentrated video file to obtain a concentrated video frame and a relative playing time of the concentrated video frame;
根据所述相对播放时间，获取与所述浓缩视频帧的所有运动目标相关联的标注信息包；Obtaining, according to the relative play time, an annotation information packet associated with all moving targets of the concentrated video frame;
解析所述标注信息包，确定所述浓缩视频帧的运动目标的标注信息；Parsing the annotation information packet to determine annotation information of a moving target of the concentrated video frame;
根据所述相对播放时间，将所述浓缩视频帧以及该浓缩视频帧的运动目 标的标注信息进行叠加显示处理，完成播放。And concentrating the concentrated video frame and the concentrated video frame according to the relative play time The labeling information of the label is superimposed and displayed to complete the playback.
所述运动目标的标注信息包括：运动目标的坐标、运动目标的高度和运动目标的宽度。The annotation information of the moving target includes: coordinates of the moving target, height of the moving target, and width of the moving target.
本发明又提供了一种浓缩视频的运动目标标注装置，包括：The invention further provides a moving target marking device for concentrating video, comprising:
获取模块，设置为获取一浓缩视频帧中的所有运动目标的标注信息以及该浓缩视频帧的相对播放时间；An obtaining module, configured to obtain annotation information of all moving targets in a concentrated video frame and a relative playing time of the concentrated video frame;
封装模块，设置为将所述浓缩视频帧和该浓缩视频帧的所有运动目标的标注信息分别封装成媒体数据包和标注信息包；The encapsulation module is configured to encapsulate the enrichment video frame and the annotation information of all the moving targets of the enriched video frame into a media data packet and an annotation information packet, respectively;
关联模块，设置为根据所述浓缩视频帧的相对播放时间，建立所述浓缩视频帧的媒体数据包和标注信息包之间的关联。The association module is configured to establish an association between the media data packet of the concentrated video frame and the labeling information packet according to the relative playing time of the concentrated video frame.
可选地，还包括：Optionally, it also includes:
保存模块，用设置为将所述媒体数据包、标注信息包，以及所述媒体数据包与标注信息包之间的关联，分别保存到浓缩视频文件中，得到目的浓缩视频文件。The saving module is configured to save the media data package, the labeling information packet, and the association between the media data packet and the labeling information package to the concentrated video file to obtain the target concentrated video file.
可选地，所述获取模块包括：Optionally, the obtaining module includes:
浓缩模块，设置为对原视频文件进行浓缩处理，得到一浓缩视频帧；The concentrating module is configured to perform concentration processing on the original video file to obtain a concentrated video frame;
提取模块，设置为对所述浓缩视频帧进行运动目标分析，提取运动目标；其中，运动目标包括：在所述浓缩视频帧中没有与其他运动子目标重叠的运动子目标，以及具有重叠关系的多个运动子目标；An extraction module, configured to perform a moving target analysis on the concentrated video frame, and extract a moving target; wherein the moving target includes: a moving sub-target that does not overlap with other moving sub-objects in the concentrated video frame, and an overlapping relationship Multiple sports sub-goals;
确定模块，设置为分别获取每一个运动目标的标注信息，确定所有运动目标的标注信息。The determining module is configured to acquire the labeling information of each moving target separately, and determine the labeling information of all the moving targets.
可选地，所述封装模块至少包括：Optionally, the encapsulating module at least includes:
第一封装子模块，设置为将所述浓缩视频帧封装成第一预设格式的媒体数据包。The first encapsulation submodule is configured to encapsulate the enriched video frame into a media packet of a first preset format.
可选地，所述封装模块还包括：Optionally, the encapsulating module further includes:
第二封装子模块，设置为将该浓缩视频帧的所有运动目标的标注信息封装成第二预设格式的标注信息包。The second encapsulation submodule is configured to encapsulate the annotation information of all the moving targets of the concentrated video frame into the annotation information packet of the second preset format.
本发明再提供了一种浓缩视频的播放装置，包括：The present invention further provides a playback device for a concentrated video, comprising:
第一解析模块，设置为解析目的浓缩视频文件，得到一浓缩视频帧以及 该浓缩视频帧的相对播放时间；a first parsing module configured to parse the target concentrated video file to obtain a concentrated video frame and Relative play time of the concentrated video frame;
第一获取模块，设置为根据所述相对播放时间，获取与所述浓缩视频帧的所有运动目标相关联的标注信息包；a first acquiring module, configured to acquire, according to the relative playing time, an annotation information packet associated with all moving targets of the concentrated video frame;
第二解析模块，设置为解析所述标注信息包，确定所述浓缩视频帧的运动目标的标注信息；a second parsing module, configured to parse the label information packet, and determine labeling information of a moving target of the concentrated video frame;
播放模块，设置为根据所述相对播放时间，将所述浓缩视频帧以及该浓缩视频帧的运动目标的标注信息进行叠加显示处理，完成播放。The playing module is configured to superimpose and display the enriched video frame and the labeling information of the moving target of the concentrated video frame according to the relative playing time to complete the playing.
可选地，所述运动目标的标注信息包括：运动目标的坐标、运动目标的高度和运动目标的宽度。Optionally, the annotation information of the moving target includes: coordinates of the moving target, a height of the moving target, and a width of the moving target.
在本发明实施例的浓缩视频的运动目标标注方法中，通过将浓缩视频帧的运动目标的标注信息和该浓缩视频帧数据分别封装写入浓缩视频文件中，减少了将两者合成的步骤，提高了处理效率，而且实现了视频帧数据与标注信息的分离；同时，本发明实施例通过视频帧的相对播放时间关联该视频帧和该视频的运动目标的标注信息，使标注信息实现了方便的与视频信息保存与传送，在浓缩视频播放过程中直接实现了运动目标的标注，从而提高了标注效率，保证了播放效率。In the moving target labeling method of the condensed video according to the embodiment of the present invention, the labeling information of the moving target of the condensed video frame and the condensed video frame data are separately encapsulated and written into the condensed video file, thereby reducing the step of synthesizing the two. The processing efficiency is improved, and the separation of the video frame data and the annotation information is realized. Meanwhile, in the embodiment of the present invention, the labeling information of the video frame and the moving target of the video is associated by the relative playing time of the video frame, so that the labeling information is convenient. The storage and transmission of the video information directly realizes the labeling of the moving target in the process of concentrating the video playing, thereby improving the labeling efficiency and ensuring the playing efficiency.
附图概述BRIEF abstract
此处所说明的附图用来提供对本发明的进一步理解，构成本申请的一部分，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。在附图中：The drawings described herein are intended to provide a further understanding of the invention, and are intended to be a part of the invention. In the drawing:
图1为本发明实施例中浓缩视频的运动目标标注方法的流程图；1 is a flowchart of a method for marking a moving target of a concentrated video according to an embodiment of the present invention;
图2为本发明实施例中浓缩视频的运动目标标注方法的关联关系确定的流程图；2 is a flowchart of determining an association relationship of a moving target labeling method for a concentrated video according to an embodiment of the present invention;
图3为本发明实施例中浓缩视频的播放方法的流程图；3 is a flowchart of a method for playing a concentrated video in an embodiment of the present invention;
图4为本发明实施例中以MP4文件为例的具体标注及播放流程示意图；4 is a schematic diagram of a specific labeling and playing process of an MP4 file as an example in the embodiment of the present invention;
图5为本发明实施例中浓缩视频的运动目标标注装置的结构示意图；FIG. 5 is a schematic structural diagram of a moving target marking device for concentrating video according to an embodiment of the present invention; FIG.
图6为本发明实施例中浓缩视频的播放装置的结构示意图。 FIG. 6 is a schematic structural diagram of a device for playing back a concentrated video according to an embodiment of the present invention.
本发明的较佳实施方式Preferred embodiment of the invention
为使本发明实施例的目的、技术方案和优点更加清楚明白，下文中将结合附图对本发明的实施例进行详细说明。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互任意组合。The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that, in the case of no conflict, the features in the embodiments and the embodiments in the present application may be arbitrarily combined with each other.
本发明针对现有技术中浓缩视频技术处理性能消耗大，标注信息与浓缩视频帧不能分离或者依靠描述文件对视频帧的运动目标进行处理，但是描述文件较大，不利于保存和传输等的问题，提供一种浓缩视频的运动目标标注方法、播放方法及装置，本发明实施例通过将浓缩视频帧的运动目标的标注信息和该浓缩视频帧数据分别封装写入浓缩视频文件中，减少了将两者合成的步骤，提高了处理效率，而且实现了视频帧数据与标注信息的分离；同时，本发明实施例通过视频帧的相对播放时间关联该视频帧和该视频的运动目标的标注信息，使标注信息实现了方便的与视频信息保存与传送，在浓缩视频播放过程中直接实现了运动目标的标注，从而提高了标注效率，保证了播放效率。The present invention is directed to the prior art that the concentrated video technology has high processing performance consumption, and the annotation information cannot be separated from the concentrated video frame or the moving object of the video frame is processed by the description file, but the description file is large, which is not conducive to preservation and transmission. The invention provides a moving target labeling method, a playing method and a device for concentrating video. In the embodiment of the present invention, the labeling information of the moving target of the concentrated video frame and the concentrated video frame data are separately encapsulated and written into the concentrated video file, thereby reducing The step of synthesizing the two improves the processing efficiency, and realizes the separation of the video frame data and the annotation information. Meanwhile, the embodiment of the present invention associates the video frame with the annotation information of the moving target of the video by the relative playing time of the video frame. The annotation information is conveniently saved and transmitted with the video information, and the moving target is directly marked in the concentrated video playback process, thereby improving the labeling efficiency and ensuring the playback efficiency.
如图1所示，本发明实施例提供一种浓缩视频的运动目标标注方法，包括：As shown in FIG. 1 , an embodiment of the present invention provides a method for marking a moving target of a concentrated video, including:
步骤100，获取一浓缩视频帧中的所有运动目标的标注信息以及该浓缩视频帧的相对播放时间；Step 100: Obtain annotation information of all moving targets in a concentrated video frame and relative playing time of the concentrated video frame.
步骤101，将获得的浓缩视频帧和该浓缩视频帧的所有运动目标的标注信息分别封装成媒体数据包和标注信息包；Step 101: Encapsulate the obtained concentrated video frame and the labeling information of all moving targets of the concentrated video frame into a media data packet and an annotation information packet, respectively;
步骤102，根据获得的浓缩视频帧的相对播放时间，建立该浓缩视频帧的媒体数据包和标注信息包之间的关联。Step 102: Establish an association between the media data packet of the concentrated video frame and the labeled information packet according to the relative playing time of the obtained concentrated video frame.
本发明上述实施例中，步骤100中的运动目标是指浓缩视频帧中运动的物体，如运动的人、车辆等；对运动目标进行箭头、圆圈等标注能够帮助观看视频的人员在很短的时间内看完所有的活动目标。In the above embodiment of the present invention, the moving target in step 100 refers to a moving object in a video frame, such as a moving person, a vehicle, etc.; an arrow, a circle, or the like for the moving target can help the person watching the video in a short time. Read all the activity goals in time.
需要说明的是，本发明实施例的浓缩视频的运动目标的标注方法是在对原视频文件的浓缩过程中进行的，即将原视频文件进行浓缩处理浓缩出一帧视频帧就继续执行以下的步骤101、步骤102，直到原视频文件浓缩完成，则运动目标的标注信息的提取也完成，方便在播放浓缩视频时直接调用该标注 信息，直接进行运动目标的标注，提高了标注效率；同时，步骤101中将获得的浓缩视频帧和该浓缩视频的运动目标的标注信息分别封装，分别写入浓缩视频文件，使视频帧数据和标注信息实现了分离，节省了两个结合的处理步骤，提高了处理效率。It should be noted that the method for labeling the moving target of the condensed video in the embodiment of the present invention is performed in the process of concentrating the original video file, that is, the original video file is condensed to concentrate one frame of the video frame, and then the following steps are performed. 101, step 102, until the concentration of the original video file is completed, the extraction of the annotation information of the moving target is also completed, and the annotation is directly called when the concentrated video is played. The information directly marks the moving target, and improves the labeling efficiency; at the same time, the concentrated video frame obtained in step 101 and the labeling information of the moving target of the concentrated video are separately packaged, respectively, and the concentrated video file is written to make the video frame data and The labeling information achieves separation, saving two combined processing steps and improving processing efficiency.
进一步的，由于将浓缩视频帧和该浓缩视频帧的运动目标的标注信息分别封装，则需要根据相对播放时间建立上述浓缩视频帧和该浓缩视频帧的运动目标的标注信息之间的关联，方便在播放视频时，通过其关联关系找到对应的浓缩视频帧和该浓缩视频的运动目标的标注信息，并将其结合播放，构成完整的视频。Further, since the labeling information of the moving video frame and the moving target of the concentrated video frame are respectively encapsulated, it is necessary to establish an association between the concentrated video frame and the labeling information of the moving target of the concentrated video frame according to the relative playing time, which is convenient. When the video is played, the corresponding concentrated video frame and the annotation information of the moving target of the concentrated video are found through the association relationship, and are combined and played to form a complete video.
本发明上述实施例中，该浓缩视频的运动目标标注方法还包括：In the above embodiment of the present invention, the moving target labeling method of the concentrated video further includes:
将建立的媒体数据包、标注信息包，以及媒体数据包与标注信息包之间的关联，分别保存到浓缩视频文件中，得到目的浓缩视频文件。The established media data package, the labeled information package, and the association between the media data package and the labeled information package are respectively saved into the concentrated video file to obtain the target concentrated video file.
本发明具体实施例中，确定一个浓缩视频帧的媒体数据包、标注信息包，以及媒体数据包与标注信息包之间的关联后，继续获取下一浓缩视频帧的媒体数据包、标注信息包，以及和媒体数据包与标注信息包之间的关联，直到所有的浓缩视频帧均处理完毕，并将每帧的媒体数据包、标注信息包，以及媒体数据包与标注信息包之间的关联分别保存，得到目的浓缩视频文件。该目的浓缩视频文件即是采用本发明实施例提供的运动目标标注方法标注运动目标得到的文件。In a specific embodiment of the present invention, after determining a media data packet, an annotation information packet, and an association between the media data packet and the annotation information packet of a concentrated video frame, the media data packet and the annotation information packet of the next concentrated video frame are continuously obtained. And the association between the media packet and the annotated packet until all of the condensed video frames have been processed, and the media packets, annotated packets, and the association between the media packets and the annotated packets for each frame Save separately to get the target concentrated video file. The condensed video file of the purpose is a file obtained by labeling a moving target by using the moving target labeling method provided by the embodiment of the present invention.
需要说明的是，其关联关系写入浓缩视频文件的元数据描述部分，元数据描述部分通常用于存储视频文件中各个部分之间的联系；例如，普通视频文件中一般至少包括音频轨、视频轨等，为了在播放时确保音频与视频同步播放，需要在元数据描述部分存储音频和视频的关系，即播放视频时应同时播放哪一个对应的音频文件。It should be noted that the association relationship is written into the metadata description part of the condensed video file, and the metadata description part is generally used to store the association between the various parts in the video file; for example, the normal video file generally includes at least an audio track and a video. In order to ensure that audio and video are played synchronously during playback, it is necessary to store the relationship between audio and video in the metadata description section, that is, which corresponding audio file should be played simultaneously when playing the video.
本发明实施例中，如图2所示，步骤100具体包括：In the embodiment of the present invention, as shown in FIG. 2, step 100 specifically includes:
步骤200，对原视频文件进行浓缩处理，得到一浓缩视频帧；Step 200: Perform concentration processing on the original video file to obtain a concentrated video frame.
步骤201，对得到的浓缩视频帧进行运动目标分析，提取运动目标；其中，运动目标包括：在浓缩视频帧中没有与其他运动子目标重叠的运动子目标，以及具有重叠关系的多个运动子目标； Step 201: Perform moving object analysis on the obtained concentrated video frame to extract a moving target, where the moving target includes: a moving sub-target that does not overlap with other moving sub-objects in the concentrated video frame, and a plurality of sports sub-objects with overlapping relationship aims;
步骤202，分别获取每一个运动目标的标注信息，确定所有运动目标的标注信息。Step 202: Acquire label information of each moving target separately, and determine labeling information of all moving targets.
本发明具体实施例中，步骤200中的浓缩处理具体为：对预设数量的帧图像进行浓缩处理的具体过程可通过一浓缩算法进行具体操作，如假设该预设数量为5帧，将输入5帧原视频文件的图像，经过一浓缩算法的处理后，输出一帧视频图像，则该帧视频图像即为本发明实施例的一浓缩视频帧图像，即得到浓缩视频帧；该浓缩视频帧是上述5帧原视频文件图像的精华，通过将没有价值的视频进行了剔除，将有价值的视频进行了合并等技术操作得到浓缩视频帧。In the specific embodiment of the present invention, the concentrating process in step 200 is specifically: the specific process of concentrating the preset number of frame images may be performed by a concentrating algorithm, for example, if the preset number is 5 frames, the input will be input. The image of the original video file of the 5 frames is processed by a condensing algorithm, and a video image of the frame is output, and the video image of the frame is a condensed video frame image of the embodiment of the present invention, that is, the condensed video frame is obtained; It is the essence of the above 5 frames of the original video file image, and the condensed video frame is obtained by merging the valuable video and combining the valuable video.
其中，该预设数量可根据用户的不同需求设置，如用户需要将一个视频浓缩成10M和用户需要将同一个视频浓缩成5M就可选择不同的预设数量；当然，浓缩后的视频越小，该预设数量的值越大。The preset number can be set according to different needs of the user. For example, if the user needs to condense one video into 10M and the user needs to condense the same video into 5M, a different preset number can be selected; of course, the smaller the concentrated video is. , the larger the value of the preset number.
本发明具体实施例中，对于运动目标的确定，可以先将浓缩视频帧中出现的各个运动的物体都看成运动子目标，然后将没有与其他运动子目标重叠的运动子目标确定为运动目标；并将与其他运动子目标有重叠关系的多个运动子目标共同确定为一个运动目标。例如，在浓缩视频帧中有两个运动子目标即运动子目标A和运动子目标B，如果运动子目标A和运动子目标B在浓缩视频帧中没有发生重叠，则将运动子目标A确定为一个运动目标，并将运动子目标B确定为另一个运动目标；如果运动子目标A和运动子目标B在该浓缩视频帧中发生重叠，那么将运动子目标A和运动子目标B共同看成一个运动目标。In the specific embodiment of the present invention, for the determination of the moving target, each moving object appearing in the concentrated video frame may be regarded as a moving sub-target, and then the moving sub-target that does not overlap with other moving sub-objects is determined as the moving target. And a plurality of sports sub-goals having overlapping relationship with other sports sub-goals are collectively determined as one moving target. For example, in a concentrated video frame, there are two moving sub-targets, a moving sub-target A and a moving sub-target B. If the moving sub-target A and the moving sub-target B do not overlap in the concentrated video frame, the moving sub-target A is determined. For one moving target, and determining the moving sub-target B as another moving target; if the moving sub-target A and the moving sub-target B overlap in the concentrated video frame, then the moving sub-target A and the moving sub-target B are seen together Become a moving target.
进一步的，步骤202中针对每一个运动目标确定相同或不同的标注信息，并保存，便于后续调用。Further, in step 202, the same or different annotation information is determined for each moving target, and saved for subsequent calls.
具体的，本发明上述实施例中，所述运动目标的标注信息包括：运动目标的坐标、运动目标的高度和运动目标的宽度。Specifically, in the foregoing embodiment of the present invention, the labeling information of the moving target includes: a coordinate of the moving target, a height of the moving target, and a width of the moving target.
本发明具体实施例中，为了准确的标注一个运动目标，至少需要获取运动目标的坐标(X轴、Y轴，立体的还需要获取Z轴)、运动目标的高度以及运动目标的宽度。In the specific embodiment of the present invention, in order to accurately mark a moving target, at least the coordinates of the moving target (X-axis, Y-axis, and the Z-axis of the stereo), the height of the moving target, and the width of the moving target need to be acquired.
本发明上述实施例中，步骤101还包括：将所述浓缩视频帧封装成第一 预设格式的媒体数据包。In the above embodiment of the present invention, step 101 further includes: encapsulating the concentrated video frame into a first Media packets in a preset format.
本发明具体实施例中，媒体数据包中包括该浓缩视频帧，其封装过程主要是将其转化成符合浓缩视频文件的播放格式的帧数据，如mp4格式、rmvb格式、mtv格式、wmv格式等等，根据浓缩视频文件的格式确定其媒体数据包的格式。In the specific embodiment of the present invention, the media data packet includes the concentrated video frame, and the encapsulation process is mainly to convert the frame data into a frame format conforming to the playback format of the concentrated video file, such as mp4 format, rmvb format, mtv format, wmv format, etc. Etc., determine the format of its media packet based on the format of the condensed video file.
进一步的，本发明上述实施例中，步骤102还包括：将该浓缩视频帧的所有运动目标的标注信息封装成第二预设格式的标注信息包。其中，Further, in the foregoing embodiment of the present invention, the step 102 further includes: encapsulating the annotation information of all the moving targets of the concentrated video frame into the annotation information packets of the second preset format. among them,
标注信息包中包括该浓缩视频帧中所有运动目标的标注信息，其封装过程主要是将该标注信息打包成符合浓缩视频文件的标注信息格式的文件，如mp4格式的播放文件的标注信息轨的格式、rmvb格式的播放文件的标注信息轨的格式等等，在此不一一列举。The labeling information packet includes the labeling information of all the moving objects in the concentrated video frame, and the packaging process mainly packs the labeling information into a file conforming to the labeling information format of the concentrated video file, such as the labeling information track of the playing file in the mp4 format. The format, the format of the labeling information track of the playback file in the rmvb format, and the like are not listed here.
本发明上述具体实施例中，根据该浓缩视频帧的相对播放时间，如距预设时间点5s时播放该帧的媒体数据包和标注信息包，则确定相对播放时间处媒体数据包和标注信息包的关联关系。In the above specific embodiment of the present invention, according to the relative playing time of the concentrated video frame, if the media data packet and the labeled information packet of the frame are played when the preset time point is 5 s, the media data packet and the labeled information at the relative playing time are determined. The association of the package.
本发明具体实施例中，由于本发明的运动目标标注方法是在原视频的文件的浓缩过程进行的标注，故在每一个浓缩视频帧处理完毕时，此时的浓缩视频文件即为目的浓缩视频文件，且浓缩及对应的标注信息保存均完成，提高浓缩效率。In the specific embodiment of the present invention, since the moving target labeling method of the present invention is marked in the enrichment process of the original video file, when each concentrated video frame is processed, the concentrated video file at this time is the target concentrated video file. And the concentration and corresponding label information are saved, which improves the concentration efficiency.
为了更好的实现上述目的，如图3所示，本发明实施例还提供一种浓缩视频的播放方法，包括：In order to achieve the above purpose, as shown in FIG. 3, an embodiment of the present invention further provides a method for playing a concentrated video, including:
步骤300，解析目的浓缩视频文件，得到一浓缩视频帧以及该浓缩视频帧的相对播放时间； Step 300, parsing the target concentrated video file to obtain a concentrated video frame and a relative playing time of the concentrated video frame;
步骤301，根据所述相对播放时间，获取与所述浓缩视频帧的所有运动目标相关联的标注信息包；Step 301: Acquire an annotation information packet associated with all moving targets of the concentrated video frame according to the relative playing time;
步骤302，解析所述标注信息包，确定所述浓缩视频帧的运动目标的标注信息； Step 302, parsing the label information packet, and determining labeling information of a moving target of the concentrated video frame;
步骤303，根据所述相对播放时间，将所述浓缩视频帧以及该浓缩视频帧的运动目标的标注信息进行叠加显示处理，完成播放。Step 303: Perform superimposed display processing on the labeled information of the concentrated video frame and the moving target of the concentrated video frame according to the relative playing time, and complete the playing.
本发明上述实施例中，浓缩视频的播放过程是与本发明提供的浓缩视频 的运动目标的标注方法相对应；其实质是与本发明提供的浓缩视频的运动目标的标注方法的过程相反。具体的，所述运动目标的标注信息包括：运动目标的坐标、运动目标的高度和运动目标的宽度。In the above embodiment of the present invention, the playing process of the concentrated video is the concentrated video provided by the present invention. The method of labeling the moving object corresponds to; the essence is opposite to the process of the labeling method of the moving object of the concentrated video provided by the present invention. Specifically, the annotation information of the moving target includes: coordinates of the moving target, a height of the moving target, and a width of the moving target.
下面结合图4，假设原视频文件为MP4文件，具体描述本发明提供的运动目标标注方法以及对应浓缩视频的播放过程：Referring to FIG. 4, it is assumed that the original video file is an MP4 file, and the moving target labeling method and the corresponding concentrated video playing process provided by the present invention are specifically described:
MP4浓缩视频中运动目标标注信息的生成步骤：Steps for generating the moving target annotation information in the MP4 concentrated video:
步骤401:将MP4浓缩视频文件进行MP4解析处理，解析出一帧浓缩视频帧数据，进行浓缩算法处理；Step 401: Perform MP4 analysis processing on the MP4 concentrated video file, parse out one frame of concentrated video frame data, and perform a concentration algorithm processing;
步骤402浓缩算法处理，如果没有输出数据，则返回步骤401，继续解析文件；如果有数据输出，则输出数据；Step 402: Concentration algorithm processing, if there is no output data, returning to step 401, continuing to parse the file; if there is data output, outputting the data;
步骤403:经过浓缩处理后的视频帧数据；Step 403: The video frame data after the concentration processing;
步骤404:经过浓缩处理后的视频帧运动目标的X轴、Y轴、Heigh高度、Width宽度等标注信息；Step 404: labeling information such as an X-axis, a Y-axis, a Heigh height, and a Width width of the moving target of the video frame after the concentration processing;
步骤405:将步骤403和步骤404分别封装成MP4文件中的一个视频轨和一个信息轨，根据相对播放时间封装，以便在播放时再相对应的关联；Step 405: Encapsulate step 403 and step 404 into a video track and an information track in the MP4 file respectively, and package according to the relative play time, so as to correspondingly associate in the playing;
步骤406:将封装的视频轨和信息轨分别写MP4浓缩视频文件，同时把相对播放时间关联上述视频轨和信息轨的信息，写入MP4文件元数据描述部分。返回步骤401处理，直到解析完原始MP4文件。Step 406: Write the encapsulated video track and the information track to the MP4 concentrated video file respectively, and simultaneously write the information related to the video track and the information track with respect to the play time to the MP4 file metadata description part. Returning to step 401 processing until the original MP4 file is parsed.
MP4浓缩视频中运动目标标注信息的播放步骤包括：The playing steps of the moving target annotation information in the MP4 concentrated video include:
步骤407:解析MP4浓缩视频文件,解析出视频帧数据以及该帧相对播放时间，根据相对播放时间找到相对应的标志信息；Step 407: Parsing the MP4 concentrated video file, parsing the video frame data and the relative playing time of the frame, and finding the corresponding flag information according to the relative playing time;
步骤408:得到视频帧数据和相对播放时间；Step 408: Obtain video frame data and relative play time.
步骤409:得到信息轨X轴、Y轴、Heigh、Width标注信息以及相对播放时间；Step 409: Obtain information information of the X-axis, Y-axis, Heigh, and Width of the information track and relative play time;
步骤410:根据相对播放时间，关联视频帧数据和该帧的运动目标的标注信息；Step 410: Associate the video frame data with the labeling information of the moving target of the frame according to the relative playing time;
步骤411:播放器播放时，对该视频帧数据和帧的运动目标标注进行叠加处理；循环到步骤407处理，直到解析完MP4浓缩视频文件，播放完成。Step 411: When the player plays, the video frame data and the moving target label of the frame are superimposed; the looping to step 407 is performed until the MP4 concentrated video file is parsed, and the playing is completed.
为了更好的实现上述目的，如图5所示，本发明实施例还提供一种浓缩 视频的运动目标标注装置，包括：In order to achieve the above purpose, as shown in FIG. 5, the embodiment of the present invention further provides a concentration. Video moving target marking device, including:
获取模块10，设置为获取一浓缩视频帧中的所有运动目标的标注信息以及该浓缩视频帧的相对播放时间；The obtaining module 10 is configured to obtain annotation information of all moving targets in a concentrated video frame and a relative playing time of the concentrated video frame;
封装模块20，设置为将所述浓缩视频帧和该浓缩视频帧的所有运动目标的标注信息分别封装成媒体数据包和标注信息包；The encapsulation module 20 is configured to encapsulate the enrichment video frame and the annotation information of all moving targets of the enriched video frame into a media data packet and an annotation information packet, respectively;
关联模块30，设置为根据所述浓缩视频帧的相对播放时间，建立所述浓缩视频帧的媒体数据包和标注信息包之间的关联。The association module 30 is configured to establish an association between the media data packet of the concentrated video frame and the annotation information packet according to the relative play time of the concentrated video frame.
本发明上述实施例中，所述装置还包括保存模块(图5中未示出)，设置为将所述媒体数据包、标注信息包和所述媒体数据包与标注信息包之间的关联，分别保存到浓缩视频文件中，得到目的浓缩视频文件。In the above embodiment of the present invention, the apparatus further includes a saving module (not shown in FIG. 5) configured to associate the media data packet, the labeling information packet, and the media data packet with the labeling information packet, Save them separately to the concentrated video file to get the target concentrated video file.
本发明上述实施例中，所述获取模块10具体可以包括：In the foregoing embodiment of the present invention, the acquiring module 10 may specifically include:
浓缩模块，设置为对原视频文件进行浓缩处理，得到一浓缩视频帧；The concentrating module is configured to perform concentration processing on the original video file to obtain a concentrated video frame;
提取模块，设置为对所述浓缩视频帧进行运动目标分析，提取运动目标；其中，运动目标包括：在所述浓缩视频帧中没有与其他运动子目标重叠的运动子目标，以及具有重叠关系的多个运动子目标；An extraction module, configured to perform a moving target analysis on the concentrated video frame, and extract a moving target; wherein the moving target includes: a moving sub-target that does not overlap with other moving sub-objects in the concentrated video frame, and an overlapping relationship Multiple sports sub-goals;
确定模块，设置为分别获取每一个运动目标的标注信息，确定所有运动目标的标注信息。The determining module is configured to acquire the labeling information of each moving target separately, and determine the labeling information of all the moving targets.
本发明上述实施例中，所述封装模块20具体可以包括：In the foregoing embodiment of the present invention, the encapsulating module 20 may specifically include:
第一封装子模块，设置为将所述浓缩视频帧封装成第一预设格式的媒体数据包。The first encapsulation submodule is configured to encapsulate the enriched video frame into a media packet of a first preset format.
本发明上述实施例中，所述封装模块20还包括：In the above embodiment of the present invention, the package module 20 further includes:
第二封装子模块，设置为将该浓缩视频帧的所有运动目标的标注信息封装成第二预设格式的标注信息包。The second encapsulation submodule is configured to encapsulate the annotation information of all the moving targets of the concentrated video frame into the annotation information packet of the second preset format.
本发明实施例的浓缩视频的运动目标标注技术方案中，通过将浓缩视频帧的运动目标的标注信息和该浓缩视频帧数据分别封装写入浓缩视频文件中，减少了将两者合成的步骤，提高了处理效率，而且实现了视频帧数据与标注信息的分离；同时，本发明实施例通过视频帧的相对播放时间关联该视频帧和该视频的运动目标的标注信息，使标注信息实现了方便的与视频信息保存与传送，在浓缩视频播放过程中直接实现了运动目标的标注，从而提高 了标注效率，保证了播放效率。In the moving target labeling technical solution of the condensed video according to the embodiment of the present invention, the labeling information of the moving target of the condensed video frame and the condensed video frame data are separately encapsulated and written into the condensed video file, thereby reducing the steps of synthesizing the two. The processing efficiency is improved, and the separation of the video frame data and the annotation information is realized. Meanwhile, in the embodiment of the present invention, the labeling information of the video frame and the moving target of the video is associated by the relative playing time of the video frame, so that the labeling information is convenient. And video information storage and transmission, in the process of concentrated video playback directly achieve the labeling of moving targets, thereby improving The labeling efficiency ensures the playback efficiency.
为了更好的实现上述目的，如图6所示，本发明实施例还提供一种浓缩视频的播放装置，包括：In order to achieve the above objective, as shown in FIG. 6, the embodiment of the present invention further provides a playback device for a concentrated video, including:
第一解析模块60，设置为解析目的浓缩视频文件，得到一浓缩视频帧以及该浓缩视频帧在一预设时间轴上的相对播放时间；The first parsing module 60 is configured to parse the target concentrated video file to obtain a concentrated video frame and a relative playing time of the concentrated video frame on a preset time axis;
第一获取模块70，设置为根据所述相对播放时间，获取与所述浓缩视频帧的所有运动目标相关联的标注信息包；The first obtaining module 70 is configured to acquire, according to the relative playing time, an annotated information packet associated with all moving targets of the concentrated video frame;
第二解析模块80，设置为解析所述标注信息包，确定所述浓缩视频帧的运动目标的标注信息；a second parsing module 80, configured to parse the label information packet, and determine label information of a moving target of the concentrated video frame;
播放模块90，设置为根据所述相对播放时间，将所述浓缩视频帧以及该浓缩视频帧的运动目标的标注信息进行叠加显示处理，完成播放。The playing module 90 is configured to perform superimposed display processing on the labeling information of the moving video frame and the moving target of the concentrated video frame according to the relative playing time to complete the playing.
本发明上述实施例中，运动目标的标注信息包括：运动目标的坐标、运动目标的高度和运动目标的宽度。In the above embodiment of the present invention, the annotation information of the moving target includes: coordinates of the moving target, height of the moving target, and width of the moving target.
以上所述是本发明的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明所述原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。The above is a preferred embodiment of the present invention, and it should be noted that those skilled in the art can also make several improvements and retouchings without departing from the principles of the present invention. It should also be considered as the scope of protection of the present invention.
工业实用性Industrial applicability
本发明实施例提出的浓缩视频的运动目标标注方法、播放方法及装置，在本发明实施例中，对获取一浓缩视频帧中的所有运动目标的标注信息以及该浓缩视频帧的相对播放时间；将所述浓缩视频帧和该浓缩视频帧的所有运动目标的标注信息分别封装成媒体数据包和标注信息包；根据所述浓缩视频帧的相对播放时间，建立所述浓缩视频帧的媒体数据包和标注信息包之间的关联。通过将浓缩视频帧的运动目标的标注信息和该浓缩视频帧数据分别封装写入浓缩视频文件中，减少了将两者合成的步骤，提高了处理效率，而且实现了视频帧数据与标注信息的分离；同时，本发明实施例通过视频帧的相对播放时间关联该视频帧和该视频的运动目标的标注信息，使标注信息实现了方便的与视频信息保存与传送，在浓缩视频播放过程中直接实现了运动目标的标注，从而提高了标注效率，保证了播放效率。 In the embodiment of the present invention, the annotation information of all the moving objects in a concentrated video frame and the relative playing time of the concentrated video frame are obtained in the embodiment of the present invention; Encapsulating information of the moving video frame and all moving targets of the concentrated video frame into a media data packet and an annotation information packet respectively; establishing a media data packet of the concentrated video frame according to a relative playing time of the concentrated video frame And the association between the tagged packets. By separately encapsulating the annotation information of the moving target of the concentrated video frame and the concentrated video frame data into the concentrated video file, the steps of synthesizing the two are reduced, the processing efficiency is improved, and the video frame data and the annotation information are realized. Separating; at the same time, the embodiment of the present invention associates the video frame with the labeling information of the moving target of the video by the relative playing time of the video frame, so that the labeling information is conveniently saved and transmitted with the video information, and is directly in the concentrated video playing process. The labeling of the moving target is realized, thereby improving the labeling efficiency and ensuring the playing efficiency.

Claims (15)

  1. 一种浓缩视频的运动目标标注方法，其特征在于，包括：A method for marking a moving target of a concentrated video, comprising:
    获取一浓缩视频帧中的所有运动目标的标注信息以及该浓缩视频帧的相对播放时间；Obtaining annotation information of all moving targets in a concentrated video frame and relative playing time of the concentrated video frame;
    将所述浓缩视频帧和该浓缩视频帧的所有运动目标的标注信息分别封装成媒体数据包和标注信息包；Encapsulating the enriched video frame and the annotation information of all moving targets of the concentrated video frame into a media data packet and an annotation information packet;
    根据所述浓缩视频帧的相对播放时间，建立所述浓缩视频帧的媒体数据包和标注信息包之间的关联。Establishing an association between the media data packet of the condensed video frame and the tagged information packet according to the relative play time of the condensed video frame.
  2. 根据权利要求1所述的运动目标标注方法，其特征在于，还包括：The method of marking a moving object according to claim 1, further comprising:
    将所述媒体数据包、标注信息包，以及所述媒体数据包与标注信息包之间的关联，分别保存到浓缩视频文件中，得到目的浓缩视频文件。The media data package, the labeling information packet, and the association between the media data packet and the labeling information package are respectively saved in the concentrated video file to obtain a target concentrated video file.
  3. 根据权利要求1所述的运动目标标注方法，其特征在于，所述获取一浓缩视频帧中的所有运动目标的标注信息以及该浓缩视频帧的相对播放时间包括：The moving target labeling method according to claim 1, wherein the acquiring the annotation information of all the moving objects in a concentrated video frame and the relative playing time of the concentrated video frame comprises:
    对原视频文件进行浓缩处理，得到一所述浓缩视频帧；Concentrating the original video file to obtain a concentrated video frame;
    对所述浓缩视频帧进行运动目标分析，提取运动目标；其中，运动目标包括：在所述浓缩视频帧中没有与其他运动子目标重叠的运动子目标，以及具有重叠关系的多个运动子目标；Performing a moving target analysis on the concentrated video frame to extract a moving target; wherein the moving target includes: a moving sub-target overlapping with other moving sub-objects in the concentrated video frame, and a plurality of moving sub-objects having overlapping relationships ;
    分别获取每一个运动目标的标注信息，确定所述所有运动目标的标注信息。Obtaining the annotation information of each moving target separately, and determining the labeling information of all the moving targets.
  4. 根据权利要求1～3任一项所述的运动目标标注方法，其特征在于，所述运动目标的标注信息包括：运动目标的坐标、运动目标的高度和运动目标的宽度。The moving target labeling method according to any one of claims 1 to 3, wherein the labeling information of the moving object comprises: a coordinate of the moving target, a height of the moving target, and a width of the moving target.
  5. 根据权利要求1所述的运动目标标注方法，其特征在于，所述将所述浓缩视频帧封装成媒体数据包包括：将所述浓缩视频帧封装成第一预设格式的媒体数据包。The method according to claim 1, wherein the encapsulating the condensed video frame into a media data packet comprises: packaging the condensed video frame into a media data packet in a first preset format.
  6. 根据权利要求1所述的运动目标标注方法，其特征在于，所述将该浓缩视频帧的所有运动目标的标注信息封装成标注信息包包括： The method for marking a moving object according to claim 1, wherein the encapsulating the annotation information of all moving targets of the concentrated video frame into the annotation information package comprises:
    将所述浓缩视频帧的所有运动目标的标注信息封装成第二预设格式的标注信息包。And labeling information of all moving targets of the concentrated video frame into an annotation information packet of a second preset format.
  7. 一种浓缩视频的播放方法，其特征在于，包括：A method for playing a concentrated video, comprising:
    解析目的浓缩视频文件，得到一浓缩视频帧以及该浓缩视频帧的相对播放时间；Parsing the target concentrated video file to obtain a concentrated video frame and a relative playing time of the concentrated video frame;
    根据所述相对播放时间，获取与所述浓缩视频帧的所有运动目标相关联的标注信息包；Obtaining, according to the relative play time, an annotation information packet associated with all moving targets of the concentrated video frame;
    解析所述标注信息包，确定所述浓缩视频帧的运动目标的标注信息；Parsing the annotation information packet to determine annotation information of a moving target of the concentrated video frame;
    根据所述相对播放时间，将所述浓缩视频帧以及该浓缩视频帧的运动目标的标注信息进行叠加显示处理，完成播放。And according to the relative playing time, the concentrated video frame and the labeling information of the moving target of the concentrated video frame are superimposed and displayed, and the playing is completed.
  8. 根据权利要求7所述的播放方法，其特征在于，所述运动目标的标注信息包括：运动目标的坐标、运动目标的高度和运动目标的宽度。The playing method according to claim 7, wherein the annotation information of the moving target comprises: a coordinate of the moving target, a height of the moving target, and a width of the moving target.
  9. 一种浓缩视频的运动目标标注装置，其特征在于，包括：A moving target marking device for concentrating video, comprising:
    获取模块，设置为获取一浓缩视频帧中的所有运动目标的标注信息以及该浓缩视频帧的相对播放时间；An obtaining module, configured to obtain annotation information of all moving targets in a concentrated video frame and a relative playing time of the concentrated video frame;
    封装模块，设置为将所述浓缩视频帧和该浓缩视频帧的所有运动目标的标注信息分别封装成媒体数据包和标注信息包；The encapsulation module is configured to encapsulate the enrichment video frame and the annotation information of all the moving targets of the enriched video frame into a media data packet and an annotation information packet, respectively;
    关联模块，设置为根据所述浓缩视频帧的相对播放时间，建立所述浓缩视频帧的媒体数据包和标注信息包之间的关联。The association module is configured to establish an association between the media data packet of the concentrated video frame and the labeling information packet according to the relative playing time of the concentrated video frame.
  10. 根据权利要求9所述的运动目标标注装置，其特征在于，还包括：The moving object marking device according to claim 9, further comprising:
    保存模块，用设置为将所述媒体数据包、标注信息包，以及所述媒体数据包与标注信息包之间的关联，分别保存到浓缩视频文件中，得到目的浓缩视频文件。The saving module is configured to save the media data package, the labeling information packet, and the association between the media data packet and the labeling information package to the concentrated video file to obtain the target concentrated video file.
  11. 根据权利要求9所述的运动目标标注装置，其特征在于，所述获取模块包括：The moving object marking device according to claim 9, wherein the obtaining module comprises:
    浓缩模块，设置为对原视频文件进行浓缩处理，得到一浓缩视频帧；The concentrating module is configured to perform concentration processing on the original video file to obtain a concentrated video frame;
    提取模块，设置为对所述浓缩视频帧进行运动目标分析，提取运动目标；其中，运动目标包括：在所述浓缩视频帧中没有与其他运动子目标重叠的运动子目标，以及具有重叠关系的多个运动子目标； An extraction module, configured to perform a moving target analysis on the concentrated video frame, and extract a moving target; wherein the moving target includes: a moving sub-target that does not overlap with other moving sub-objects in the concentrated video frame, and an overlapping relationship Multiple sports sub-goals;
    确定模块，设置为分别获取每一个运动目标的标注信息，确定所有运动目标的标注信息。The determining module is configured to acquire the labeling information of each moving target separately, and determine the labeling information of all the moving targets.
  12. 根据权利要求9所述的运动目标标注装置，其特征在于，所述封装模块至少包括：The moving object marking device according to claim 9, wherein the packaging module comprises at least:
    第一封装子模块，设置为将所述浓缩视频帧封装成第一预设格式的媒体数据包。The first encapsulation submodule is configured to encapsulate the enriched video frame into a media packet of a first preset format.
  13. 根据权利要求12所述的运动目标标注装置，其特征在于，所述封装模块还包括：The moving object marking device according to claim 12, wherein the packaging module further comprises:
    第二封装子模块，设置为将该浓缩视频帧的所有运动目标的标注信息封装成第二预设格式的标注信息包。The second encapsulation submodule is configured to encapsulate the annotation information of all the moving targets of the concentrated video frame into the annotation information packet of the second preset format.
  14. 一种浓缩视频的播放装置，其特征在于，包括：A playback device for a concentrated video, comprising:
    第一解析模块，设置为解析目的浓缩视频文件，得到一浓缩视频帧以及该浓缩视频帧的相对播放时间；a first parsing module configured to parse the target concentrated video file to obtain a concentrated video frame and a relative playing time of the concentrated video frame;
    第一获取模块，设置为根据所述相对播放时间，获取与所述浓缩视频帧的所有运动目标相关联的标注信息包；a first acquiring module, configured to acquire, according to the relative playing time, an annotation information packet associated with all moving targets of the concentrated video frame;
    第二解析模块，设置为解析所述标注信息包，确定所述浓缩视频帧的运动目标的标注信息；a second parsing module, configured to parse the label information packet, and determine labeling information of a moving target of the concentrated video frame;
    播放模块，设置为根据所述相对播放时间，将所述浓缩视频帧以及该浓缩视频帧的运动目标的标注信息进行叠加显示处理，完成播放。The playing module is configured to superimpose and display the enriched video frame and the labeling information of the moving target of the concentrated video frame according to the relative playing time to complete the playing.
  15. 根据权利要求14所述的播放装置，其特征在于，所述运动目标的标注信息包括：运动目标的坐标、运动目标的高度和运动目标的宽度。 The playback apparatus according to claim 14, wherein the annotation information of the moving object comprises: coordinates of the moving target, a height of the moving target, and a width of the moving target.
PCT/CN2015/072793 2014-07-28 2015-02-11 Labelling method for moving objects of concentrated video, and playing method and device WO2015117572A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410364565.8 2014-07-28
CN201410364565.8A CN105323501A (en) 2014-07-28 2014-07-28 Concentrated video moving object marking method, playing method and apparatus thereof

Publications (1)

Publication Number Publication Date
WO2015117572A1 true WO2015117572A1 (en) 2015-08-13

Family

ID=53777348

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/072793 WO2015117572A1 (en) 2014-07-28 2015-02-11 Labelling method for moving objects of concentrated video, and playing method and device

Country Status (2)

Country Link
CN (1) CN105323501A (en)
WO (1) WO2015117572A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110738709A (en) * 2019-09-10 2020-01-31 北京中盾安全技术开发公司 video evaluation method based on two-dimensional code and video evaluation system thereof
CN114449316A (en) * 2021-12-02 2022-05-06 北京快乐茄信息技术有限公司 Video processing method and device, electronic equipment and storage medium

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113038265B (en) * 2021-03-01 2022-09-20 创新奇智(北京)科技有限公司 Video annotation processing method and device, electronic equipment and storage medium
CN113949823A (en) * 2021-09-30 2022-01-18 广西中科曙光云计算有限公司 Video concentration method and device

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012019417A1 (en) * 2010-08-10 2012-02-16 中国科学院自动化研究所 Device, system and method for online video condensation
CN103077227A (en) * 2012-12-31 2013-05-01 浙江元亨通信技术股份有限公司 Video concentration retrieval analysis method and system thereof
CN103106250A (en) * 2013-01-14 2013-05-15 浙江元亨通信技术股份有限公司 Intelligent analysis and retrieval method for video surveillance and system thereof
CN103617234A (en) * 2013-11-26 2014-03-05 公安部第三研究所 Device and method for active video concentration
CN103685975A (en) * 2012-09-05 2014-03-26 中兴通讯股份有限公司 Video playing system and method

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102010031429A1 (en) * 2010-07-16 2012-01-19 Robert Bosch Gmbh Method for providing a combination video
CN101930779B (en) * 2010-07-29 2012-02-29 华为终端有限公司 Video commenting method and video player
CN103345492A (en) * 2013-06-25 2013-10-09 无锡赛思汇智科技有限公司 Method and system for video enrichment

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012019417A1 (en) * 2010-08-10 2012-02-16 中国科学院自动化研究所 Device, system and method for online video condensation
CN103685975A (en) * 2012-09-05 2014-03-26 中兴通讯股份有限公司 Video playing system and method
CN103077227A (en) * 2012-12-31 2013-05-01 浙江元亨通信技术股份有限公司 Video concentration retrieval analysis method and system thereof
CN103106250A (en) * 2013-01-14 2013-05-15 浙江元亨通信技术股份有限公司 Intelligent analysis and retrieval method for video surveillance and system thereof
CN103617234A (en) * 2013-11-26 2014-03-05 公安部第三研究所 Device and method for active video concentration

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110738709A (en) * 2019-09-10 2020-01-31 北京中盾安全技术开发公司 video evaluation method based on two-dimensional code and video evaluation system thereof
CN114449316A (en) * 2021-12-02 2022-05-06 北京快乐茄信息技术有限公司 Video processing method and device, electronic equipment and storage medium
CN114449316B (en) * 2021-12-02 2023-09-22 北京快乐茄信息技术有限公司 Video processing method and device, electronic equipment and storage medium

Also Published As

Publication number Publication date
CN105323501A (en) 2016-02-10

Similar Documents

Publication Publication Date Title
WO2019223361A1 (en) Video analysis method and apparatus
JP2011529286A5 (en)
KR101887548B1 (en) Method and apparatus of processing media file for augmented reality services
US20150193970A1 (en) Video playing method and system based on augmented reality technology and mobile terminal
RU2011107253A (en) RECORDING MEDIA, PLAYBACK DEVICE AND INTEGRAL DIAGRAM
WO2015117572A1 (en) Labelling method for moving objects of concentrated video, and playing method and device
RU2014138631A (en) AUTOMATIC DIGITAL ASSEMBLY AND LABELING OF DYNAMIC VIDEO IMAGES
RU2011106942A (en) PROCESSING 3D SUBTITLE DISPLAY
CN101828351B (en) Apparatus and method for storing and reading a file having a media data container and a metadata container
US20130282715A1 (en) Method and apparatus of providing media file for augmented reality service
US20160379410A1 (en) Enhanced augmented reality multimedia system
RU2012151489A (en) SIMULATED VIDEO WITH ADDITIONAL VIEWPOINTS AND AN ENHANCED RESOLUTION CAPABILITY FOR TRANSPORT MOTION SURVEILLANCE CAMERAS
US10412395B2 (en) Real time frame alignment in video data
US20180048877A1 (en) File format for indication of video content
CN105979349A (en) Audio frequency data processing method and device
CN101335591B (en) Apparatus and method for processing a bitstream
CN113515997A (en) Video data processing method and device and readable storage medium
RU2016135266A (en) METHOD AND DEVICE FOR PLAYING SUBTITLES 3D VIDEO
CN103873804B (en) Video replay time axis and content synchronous control method for embedded NVR
US9426403B2 (en) Video playback system and method
US20100278517A1 (en) Video decoding device
EP4096227A1 (en) Coordinates as ancillary data
JP2015082692A (en) Video editing device, video editing method, and video editing program
WO2017004933A1 (en) Video recording system for synchronously integrating speed information into video in real time
CN115810209A (en) Speaker recognition method and device based on multi-mode feature fusion network

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: 15746072

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15746072

Country of ref document: EP

Kind code of ref document: A1