WO2016065972A1 - 产生视频帧集合的方法、设备及服务器 - Google Patents

产生视频帧集合的方法、设备及服务器 Download PDF

Info

Publication number
WO2016065972A1
WO2016065972A1 PCT/CN2015/086493 CN2015086493W WO2016065972A1 WO 2016065972 A1 WO2016065972 A1 WO 2016065972A1 CN 2015086493 W CN2015086493 W CN 2015086493W WO 2016065972 A1 WO2016065972 A1 WO 2016065972A1
Authority
WO
WIPO (PCT)
Prior art keywords
video
densely
played
frames
video frames
Prior art date
Application number
PCT/CN2015/086493
Other languages
English (en)
French (fr)
Inventor
梁捷
Original Assignee
广州市动景计算机科技有限公司
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
Family has litigation
First worldwide family litigation filed litigation Critical https://patents.darts-ip.com/?family=55856568&utm_source=google_patent&utm_medium=platform_link&utm_campaign=public_patent_search&patent=WO2016065972(A1) "Global patent litigation dataset” by Darts-ip is licensed under a Creative Commons Attribution 4.0 International License.
Application filed by 广州市动景计算机科技有限公司 filed Critical 广州市动景计算机科技有限公司
Priority to US15/522,546 priority Critical patent/US10313712B2/en
Publication of WO2016065972A1 publication Critical patent/WO2016065972A1/zh

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]
    • H04N21/20—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
    • H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
    • H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
    • 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
    • H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
    • H04N21/23424—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving splicing one content stream with another content stream, e.g. for inserting or substituting an advertisement
    • 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
    • H04N21/238—Interfacing the downstream path of the transmission network, e.g. adapting the transmission rate of a video stream to network bandwidth; Processing of multiplex streams
    • H04N21/2387—Stream processing in response to a playback request from an end-user, e.g. for trick-play
    • 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
    • H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests
    • H04N21/2402—Monitoring of the downstream path of the transmission network, e.g. bandwidth available
    • 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
    • H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests
    • H04N21/2407—Monitoring of transmitted content, e.g. distribution time, number of downloads
    • 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]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/434—Disassembling of a multiplex stream, e.g. demultiplexing audio and video streams, extraction of additional data from a video stream; Remultiplexing of multiplex streams; Extraction or processing of SI; Disassembling of packetised elementary stream
    • H04N21/4343—Extraction or processing of packetized elementary streams [PES]
    • 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]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
    • 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
    • H04N21/85—Assembly of content; Generation of multimedia applications
    • H04N21/854—Content authoring
    • H04N21/8549—Creating video summaries, e.g. movie trailer

Definitions

  • the present invention relates to the field of Internet, and in particular, to a method, device and server for generating a set of video frames.
  • the video of a general variety show often lasts for more than 2 hours. If it is a TV series, the playing time will be longer.
  • some video resource providers provide exciting videos (a collection of wonderful video frames), such as a wonderful preview of a movie, a wonderful preview of a certain TV series, etc., to attract users' attention. Based on their own understanding and understanding, the video resource provider extracts the wonderful video clips they think from the entire video resources, and then synthesizes such wonderful videos.
  • a technical problem to be solved by the present invention is to provide a method, device and server for generating a video frame set, which can make the majority of users in the original video (also referred to as "original video") think more exciting video content. Stitch into a collection of video frames.
  • a method of generating a set of video frames including: statistical viewing Selecting a density of each video frame in the frequency; selecting at least one selected video frame from the video according to the playing density, wherein the playing density of the video frame is a ratio of the number of times the video frame is played to the number of times the video is played And splicing the selected video frames to form a video frame set.
  • the method may further comprise: recording a time stamp of the selected video frame in the video.
  • the step of selecting at least one selected video frame from the video comprises: filtering out the densely played video frames whose playing density is greater than a predetermined threshold, and constructing the filtered densely played video frames to form at least one densely played video segment;
  • the selected video frame is extracted from a densely played video segment; wherein the consecutive densely played video frames constitute a densely played video segment.
  • the step of selecting at least one selected video frame from the video may further comprise determining the number of the selected video frames to be extracted in each of the densely played video segments.
  • the step of determining the number of selected video frames to be extracted in each of the densely played video segments comprises: calculating an average play density of all video frames in each of the densely played video segments; calculating each of the densely played video segments The proportion of video frames whose playback intensity is greater than or equal to the average play density; and the number of selected video frames to be extracted in each densely played video segment according to the scale.
  • the determining, according to the ratio, the number of the selected video frames extracted in each of the densely played video segments comprises: stepping down from each of the intensive according to the ratio from large to small Plays the number of selected video frames extracted from the video clip.
  • the number of the selected video frames extracted in the densely played video clips with a larger proportion is greater than or equal to the number of the selected video frames extracted in the densely played video clips.
  • the number of actual video frames in the densely played video segment is less than the determined number of selected video frames to be extracted in the densely played video segment, all video frames in the densely played video segment are extracted.
  • the step of determining the number of the selected video frames to be extracted in each of the densely played video segments comprises: determining the length of each of the densely played video segments, that is, the number of video frames included in each of the densely played video segments And determining the number of the selected video frames to be extracted in each of the densely played video segments based on the length of each of the densely played video segments.
  • the step of determining, according to the length of each of the densely played video segments, the number of selected video frames extracted in each of the densely played video segments comprises: according to each of the densely played video frames
  • the length of the frequency segment is from long to short, and the number of selected video frames extracted in each of the densely played video segments is gradually reduced.
  • the number of the selected video frames extracted in the long-length densely played video segment is greater than or equal to the number of the selected video frames extracted in the short-length densely played video segment.
  • all video frames in the densely played video segment are extracted, and Describe all video frames as the number of the selected selected video frames;
  • the step of splicing the selected video frames to form the video frame set comprises: arranging the extracted selected video frames step by step according to more or less; wherein, the selected video in the same densely played video segment The frames are arranged in the order in which they are played in the video.
  • a predetermined number of video frames are extracted in each of the densely played video segments; or all video frames are extracted in each of the densely played video segments .
  • the selected video frame is randomly extracted from each of the densely played video segments; or the most densely populated video of each densely played video segment is extracted Several video frames before the frame are used as the selected video frame.
  • the step of selecting at least one selected video frame from the video comprises selecting a predetermined predetermined number of video frames of the most densely populated video frame as the selected video frame.
  • the step of selecting a video frame in the splicing to form a video frame set comprises splicing the selected video frames in sequence according to a play time sequence of the selected video frames in the video to form the video frame set.
  • an apparatus for generating a video frame set including: a statistical device, configured to count play density of each video frame in a video; and a video frame selecting device, configured to: according to each video frame in the video Selecting at least one selected video frame from the video, wherein the playing density of the video frame is the ratio of the number of times the video frame is played to the number of times the video is played; and the splicing device is configured to splicing the selected video frame to form a video frame. set.
  • the device may further comprise: a statistical device for counting play density; and/or a time stamp recording device for recording a time stamp of the selected video frame in the video.
  • the video frame selection device may include: a video frame screening device, configured to filter the densely played video frames whose playback density is greater than a predetermined threshold, and form the filtered densely played video frames to form at least one densely played video segment;
  • the successive densely played video frames constitute a densely played video segment; and
  • the video frame extracting means is configured to extract the selected video frame from the at least one densely played video segment.
  • the video frame selecting means may further comprise: frame number determining means for determining the number of the selected video frames to be extracted in each of the densely played video segments.
  • the frame number determining means may comprise: an average play density calculation means for calculating an average play density of all video frames in each of the densely played video segments; and a ratio calculating means for calculating each of the densely played video segments a ratio of a video frame whose playback intensity is greater than or equal to an average play density; and a first frame number determining means for determining, according to a ratio, a number of the selected video frames to be extracted in each of the densely played video segments.
  • the first frame number determining means is specifically configured to gradually reduce the number of the selected video frames extracted in each of the densely played video segments according to the ratio from large to small. That is, more selected video frames are extracted from the densely played video clips, and fewer selected video frames are extracted from the smaller proportion of densely played video clips.
  • the frame number determining means may comprise: video segment length determining means for determining a length of each densely played video segment, that is, a number of video frames included in each densely played video segment; and a second frame number And determining means for determining the number of the selected video frames to be extracted in each of the densely played video segments according to the length of the densely played video segment.
  • the second frame number determining means is configured to gradually reduce the number of the selected video frames extracted in each of the densely played video segments according to the length of each of the densely played video segments from long to short. number. In the densely played video clips with longer lengths, more selected video frames are extracted, and in the densely played video clips with shorter lengths, fewer selected video frames are extracted.
  • a server comprising:
  • a transceiver for counting the intensity of playback of each video frame in the video
  • a processor configured to select at least one selected video frame from the video according to the play density, wherein a play density of the video frame is a ratio of a number of times the video frame is played to a number of times the video is played And splicing the selected video frames to form the set of video frames.
  • the selecting, by the processor, the at least one selected video frame from the video comprises: screening the play secret a densely-played video frame whose degree is greater than a predetermined threshold; the filtered densely-played video frame constitutes at least one densely-played video segment; and the selected video frame is extracted from the at least one densely-played video segment; wherein, the continuous Densely playing video frames constitutes a densely played video clip.
  • the selecting, by the processor, the at least one selected video frame from the video further comprises determining the number of the selected video frames extracted in each of the densely played video segments.
  • the processor determines a number of selected video frames extracted in each of the densely played video segments, including: calculating an average play density of all video frames in each of the densely played video segments; calculating each of the dense And playing a percentage of the video frames in the video clip whose play density is greater than or equal to the average play density; and determining the number of the selected video frames extracted in each of the densely played video segments according to the ratio.
  • the processor determines, according to the ratio, the number of selected video frames extracted in each of the densely played video segments, including: gradually reducing each of the plurality according to the ratio from large to small The number of selected video frames extracted in the video clip.
  • said processor determines a number of selected video frames extracted in each of the densely played video segments, comprising: determining a length of each of the densely played video segments; and based on a length of each of the densely played video segments, Determining the number of selected video frames extracted in each of the densely played video segments.
  • the determining, by the processor, the number of the selected video frames extracted in each of the densely played video segments according to the length of each of the densely played video segments comprises: according to each of the densely played video segments The length is from long to short, and the number of selected video frames extracted in each of the densely played video segments is gradually reduced.
  • the embodiment of the invention further provides a computer readable storage medium comprising computer execution instructions for the computer to execute all or part of the steps of the method for generating a video frame set when the processor of the computer executes the computer execution instructions.
  • the invention selects the selected video frame according to the dense playing degree of the video frames in the video, and splices the selected video frames together to form a video frame set, which can splicing the wonderful video content into the video frame. In the collection.
  • FIG. 1 is a flow chart of a method of generating a set of video frames in accordance with a first embodiment of the present invention.
  • FIG. 2 is a flow chart of a method of generating a set of video frames in accordance with a second embodiment of the present invention.
  • step S220 of FIG. 4 is a flow chart of an implementation of step S220 of FIG.
  • FIG. 5 is a flow chart of another implementation of step S220 in FIG.
  • Figure 6 is a block diagram of an apparatus for generating a set of video frames in accordance with a third embodiment of the present invention.
  • Figure 7 is a block diagram of an apparatus for generating a set of video frames in accordance with a fourth embodiment of the present invention.
  • FIGS. 6 and 7 are block diagram of sub-devices that can be included in the video frame selection device 200 of FIGS. 6 and 7.
  • FIG. 9 is a block diagram showing an implementation of the frame number determining means 220 of FIG.
  • FIG. 10 is a block diagram showing another implementation of the frame number determining means 220 in FIG.
  • FIG. 11 is a schematic structural diagram of a server according to an embodiment of the present invention.
  • the wonderful video formed by the program can reflect the wonderful video of the user's perspective.
  • the formed wonderful video can be changed in real time according to the user's playing situation, and the user can view the most exciting video to other users, so that the user can view what most users think.
  • Wonderful video
  • FIG. 1 is a flow chart of a method of generating a set of video frames in accordance with a first embodiment of the present invention.
  • step S100 the playback intensity of each video frame in the video is counted.
  • the play density refers to the ratio of the number of times the video frame is played to the number of times the video is viewed. For example, if the number of times the video frame with the play time stamp is 1:00:00 is 80, and the number of times the video is played is 100, the playback intensity of the video frame is 80%.
  • This embodiment takes one mode as an example, and specifically includes:
  • the video play server records which video frames the user viewed, and records that the video is played once, and the viewed video frames are played once.
  • the play information may include: the number of times of playing each video frame, and the number of times the video is played. It is considered to be the number of times the video connection is clicked for playback) and/or the time stamp of each video frame, and so on.
  • the play density of each video frame is calculated according to the statistical play information, thereby obtaining a ratio of the number of times the video frame is played to the number of times the video is played.
  • the intensity of the broadcast can be counted by the video provider itself, or can be entrusted to other institutions for statistics, or directly from other databases.
  • the first-hand play record is often grasped, and such statistics can be realized more accurately and quickly. Playback intensity can be continuously updated as more users use it.
  • step S200 at least one "selected video frame” is selected from the video according to the intensity of playback of each video frame in the video.
  • selected video frame refers to a video frame selected from the original video to be placed in the video frame set.
  • the "play density" of a video frame is the ratio of the number of times the video frame is played to the number of times the video is played.
  • the user may instruct fast forward, may jump, or end early, not every video frame will be played. Therefore, some video frames are played more often, and some video frames are played less frequently.
  • By counting the behavior of a large number of users it can be determined which video frames are more interesting to the user. Therefore, by selecting "selected video frames" according to the degree of play density of each video frame, it is more likely to pick out content that is more interesting to the user.
  • the degree of intensive playback of video frames can be counted by the video provider itself, or it can be commissioned by a specialized survey company or other organization, or it can be obtained directly from other databases.
  • step S400 After selecting the "selected video frames", in step S400, these "selected video frames” are stitched together to form a video frame set.
  • the "selected video frames” can be spliced in sequence in the original video.
  • select video frames is to consider the degree of dense playback of video frames, you can pick out more Video frames that have been viewed by multiple users are more likely to pick out content that users prefer.
  • FIG. 2 is a flow chart of a method of generating a set of video frames in accordance with a second embodiment of the present invention.
  • the method shown in FIG. 2 has one more step than the method shown in FIG. 1.
  • Step S300, the other steps are detailed in FIG. 1, and details are not described herein again.
  • step S100 of the method shown in FIG. 2 the playing density of each video frame in the video is first counted, and the process of the statistics is as described above, and details are not described herein again.
  • the intensity of playback can be continuously updated with the use of more users.
  • the method can perform subsequent steps after counting the usage records of enough users. It is also possible to perform subsequent steps after counting a certain number of user usage records to select "selected video frames" and form a video frame set; then use a new playback intensity after further counting a certain number of user usage records. , re-select "Select Video Frame” to re-form a new video frame set to continuously optimize the content of the video frame set.
  • step S200 at least one "selected video frame" is selected from the video based on the degree of play density of each video frame in the statistical video.
  • step S300 the time stamp of each "selected video frame" in the original video can be further recorded.
  • the time stamp of each "selected video frame" in the original video can be further recorded.
  • the timestamps of all video frames can be pre-recorded, and each video frame corresponds to its timestamp. In this way, it is not necessary to specifically record the time stamp of the "selected video frame".
  • a certain video frame can be specified, and its time stamp in the original video can be determined, so that the video position in the original video can be determined.
  • step S200 the time stamp of the "selected video frame” and the identification code of the "selected video frame” are specifically recorded correspondingly.
  • these "selected video frames" are spliced as described with reference to FIG. 1, thereby forming a set of video frames.
  • step S400 After the "selected video frames" are selected, these “selected video frames” are spliced in step S400, thereby forming a video frame set.
  • the "selected video frames” can be spliced in sequence in the original video.
  • selecting the "selected video frame" is to consider the dense playback degree of the video frame, it is possible to select a video frame that has been viewed by more users, and thus it is more likely to select content that the majority of users prefer.
  • step S400 and step S300 may be reversed or may be performed simultaneously.
  • step S200 in FIGS. 1 and 2 will be further described.
  • step S200 the play density of each video frame can be sorted. Then, the first predetermined number of video frames with the strongest playback density are directly selected as the "selected video frames". The predetermined number can be determined as needed.
  • the “selected video frame” is a number of video frames with the most playback times, which is relatively easy to reflect which video frames the user pays attention to.
  • a predetermined number of video frames having the highest number of plays are in consecutive video segments, such that the video frame set may not reflect more content in the video.
  • step S200 The method of performing step S200 described above with reference to FIGS. 1 and 2 by densely playing a video clip is further described below with reference to FIGS. 3 through 5.
  • step S210 a video frame in which the broadcast density is greater than a predetermined threshold is selected, which is referred to herein as a "dense play video frame.”
  • the predetermined threshold here may be the same value pre-assigned for all videos, for example 60%.
  • the predetermined threshold may also vary for different videos.
  • this predetermined threshold can also be determined according to the actual playing condition of each video frame in the video. For example, it can be set to a certain proportion of the highest playback intensity, for example, half of the highest playback intensity, two-thirds, and the like.
  • the densely played video frames thus filtered have some consecutive densely played video frames, and these consecutive densely played video frames constitute a video segment, which is referred to herein as a "dense video clip.” Whether multiple video frames are continuous can be judged according to their timestamps. When a video frame is not connected to other video frames, it can also be considered that the video frame separately constitutes a video segment, except that the video segment has only one video frame.
  • the plurality of densely played video frames that are filtered out will constitute at least one densely played video segment.
  • step S220 the number of "selected video frames" to be extracted in each of the densely played video segments is determined. This step is an optional step.
  • step S230 a "selected video frame" is extracted from the at least one densely played video segment.
  • step S230 when the "selected video frame" is extracted from the densely played video segment, the “selected video frame” may be randomly extracted from the densely played video segment, or the most densely-previous plurality of the densely-played video segments may be extracted. Video frames are used as "selected video frames.”
  • step S230 the "selected video frames" extracted in each of the densely played video segments may be continuous or discontinuous. If successive "selected video frames" are extracted, the user can be presented with relatively small video segments.
  • the specific number of extractions may be determined by the above step S220, or may not be determined by the above step S220.
  • FIG. 3 shows a case including a step of determining the number of "selected video frames" to be extracted in each densely played video segment.
  • step S230 it is also possible to predetermine how many "selected video frames" to be extracted in each video frame in step S230 without going through step S220.
  • the same number (predetermined number) of video frames are extracted in each of the densely played video segments, for example, five video frames are extracted in each of the densely played video segments.
  • step S210 when the predetermined threshold set in step S210 is high, it is possible that the number of densely played video frames that are filtered out is small, so that all video frames can also be extracted in each of the densely played video segments.
  • step S220 is not required, but a predetermined number of "selected video frames" can be directly extracted in step S230.
  • step S220 determining the number of video frames to be extracted in each of the densely played video segments according to the actual video playback statistics can better reflect the user's perspective.
  • step S220 of FIG. 4 is a flow chart of an implementation of step S220 of FIG.
  • how many video frames to be extracted in the densely played video segment is determined according to the distribution of the broadcast density of video frames in the densely played video segment.
  • step S222 the average play density of all video frames in each of the densely played video segments is calculated.
  • step S223 the proportion of the video frames in which the density of the intensively played video is greater than or equal to the average playback intensity is calculated.
  • this ratio is small, it is likely that only a few videos in this video clip are played more than once due to some accidental factors such as fast forward. Therefore, it can be considered that this video clip is less important and can extract fewer video frames from this video clip.
  • step S224 the number of "selected video frames" to be extracted in each densely played video segment may be determined according to the ratio calculated in step S223.
  • more "selected video frames” can be extracted from the densely-played video segments with larger ratios, and fewer “selected video frames” are extracted from the densely-played video segments with smaller ratios.
  • the number of extracted "selected video frames" can be gradually reduced in descending order of the above ratio from high to low (or large to small).
  • the video clip 1 has the highest ratio of 90%, and the video clip 2 has the highest ratio.
  • 80%, ..., 5 frames are extracted from video segment 1, and 4 frames are extracted from video segment 2.
  • consecutive frames may be extracted, or discontinuous frames may be extracted; frames may be randomly extracted, and several frames with the strongest broadcast density may be extracted.
  • the number of actual video frames in a densely played video segment is smaller than that determined in accordance with the above ratio in the densely played video segment.
  • the number of "selected video frames" extracted. At this point, all of the video frames in the densely played video clip can be extracted. All video frames are taken as the number of extracted "selected video frames”.
  • FIG. 5 is a flow chart of another implementation of step S220 in FIG.
  • how many video frames are to be extracted in the densely played video segment is determined according to the length of the densely played video segment (ie, the number of video frames).
  • step S226 the length of each densely played video segment is determined.
  • the length of the densely played video clip can be represented by the number of video frames contained in the densely played video clip.
  • step S2208 the number of "selected video frames" to be extracted in each densely played video segment is determined based on the length of the densely played video segment.
  • More "selected video frames" can be extracted from the long-length densely played video segments, and fewer “selected video frames” are extracted from the shorter-length densely played video segments.
  • the number of extracted "selected video frames" can be gradually reduced in descending order of the length of the densely played video segments from large to small (or long to short).
  • consecutive frames can be extracted, or non-contiguous frames can be extracted; frames can be randomly extracted, and several frames with the strongest broadcast density can be extracted.
  • step S220 Two ways in which the number of "selected video frames" to be extracted in each densely played video segment can be determined in step S220 are described above with reference to FIGS. 4 and 5.
  • step S220 In the case where the number of "selected video frames" to be extracted is determined by step S220, it can be considered that the video clip from which a plurality of "selected video frames" are selected is more received by the user.
  • the "selected video frames" extracted from the densely played video segments with more “selected video frames” extracted may be ranked first, and the densely played video segments with less “selected video frames” selected may be extracted. "Selected video frames” are arranged after. Thereby placing content that may be more of a user's attention in front of the set of video frames.
  • the splicing of the selected video frames to form the video frame set includes: arranging the extracted selected video frames step by step according to more or less;
  • the "selected video frames" in the same densely played video clip can be arranged in the order of their playing time in the video, so as to ensure the temporal relationship of the video content in a video clip.
  • the extracting the selected video frame from the at least one densely played video segment comprises: extracting a predetermined number of video frames in each of the densely played video segments; or All video frames are extracted from each densely played video clip.
  • the extracting the selected video frame from the at least one densely played video segment comprises: randomly extracting the selected video frame in each of the densely played video segments; or extracting each densely played video Several video frames preceding the most densely populated video frame are used as the selected video frame.
  • FIGS. 6 through 10 an apparatus for generating a video frame set according to the present invention, which can be used to perform the method of generating a video frame set described above with reference to FIGS. 1 through 5, is described in detail with reference to FIGS. 6 through 10.
  • Figure 6 is a block diagram of an apparatus for generating a set of video frames in accordance with a third embodiment of the present invention.
  • the apparatus shown in FIG. 6 includes a statistical device 100, a video frame selection device 200, and a splicing device 400.
  • the statistical device 100 counts the intensity of playing each video frame in the video; specifically: information recording
  • the device is configured to record playing information of each video; and the calculating device is configured to calculate a playing intensity of each video frame in a video according to the playing information.
  • the video frame selection means 200 selects at least one "selected video frame" from the video according to the degree of play density of each video frame in the video. As described above, the playback intensity of a video frame is the ratio of the number of times the video frame is played to the number of times the video is played.
  • the video frame selection device 200 can select "selected video frames" in the same manner as described above with reference to FIGS. 1 through 5.
  • the splicing device 400 splices the "selected video frames" selected by the video frame selection device 200 to form a set of video frames.
  • the splicing device 400 can splicing the "selected video frames" in the same manner as described above with reference to FIGS. 1 through 5.
  • Figure 7 is a block diagram of an apparatus for generating a set of video frames in accordance with a fourth embodiment of the present invention.
  • the apparatus shown in FIG. 7 adds a time stamp recording apparatus 300.
  • the time stamp recording device 300 records the time stamp of the "selected video frame" in the video.
  • FIGS. 6 and 7 are block diagram of sub-devices that can be included in the video frame selection device 200 of FIGS. 6 and 7.
  • the video frame selection device 200 may include a video frame screening device 210, a frame number determining device 220, and a video frame extraction device 230. Of course, it may also include only the video frame screening device 200 and the video frame extraction device 230 (not shown), that is, in order to limit the size of the highlight video, it is necessary to limit the number of extracted "selected video frames".
  • the video frame screening device 210 filters out densely played video frames that are more dense than a predetermined threshold.
  • the continuous densely played video frames constitute a densely played video segment, and the filtered densely played video frames constitute at least one densely played video segment.
  • the frame number determining means 220 determines the number of "selected video frames" to be extracted in each of the densely played video segments.
  • the video frame extraction means 230 extracts "selected video frames" from at least one of the densely played video segments.
  • the frame number determining means 220 is not necessarily required for the same reason as described above for the method according to the present invention.
  • the number of frames to be extracted in each of the densely played video segments may be predetermined, for example, a predetermined number of video frames are extracted, or all video frames are extracted.
  • the frame number determining means 220 can be realized in the following two ways.
  • FIG. 9 is a block diagram showing an implementation of the frame number determining means 220 of FIG.
  • the frame number determining means 220 may include an average play density calculating means 222, a ratio calculating means 223, and a first frame number determining means 224.
  • the average play density calculation means 222 calculates the average play density of all video frames in each of the densely played video clips.
  • the ratio calculating means 223 calculates the proportion of video frames in which each of the densely played video clips is more dense than the average play density.
  • the first frame number determining means 224 determines the number of "selected video frames" to be extracted in each of the densely played video segments in accordance with the above ratio.
  • the first frame number determining apparatus is specifically configured to gradually reduce the number of the selected video frames extracted in each of the densely played video segments according to the ratio from large to small. That is to say, more "selected video frames" are extracted in the densely-played video segments with larger ratios, and fewer "selected video frames” are extracted in the densely-played video segments with smaller proportions.
  • FIG. 10 is a block diagram showing another implementation of the frame number determining means 220 in FIG.
  • the frame number determining means 220 may include a video segment length determining means 226 and a second frame number determining means 228.
  • the video clip length determining means 226 determines the length of each densely played video clip, that is, the number of video frames included in each densely played video clip.
  • the second frame number determining means 228 determines the number of "selected video frames" to be extracted in each of the densely played video segments based on the length of the densely played video segment.
  • the second frame number determining means 228 gradually reduces the number of the selected video frames extracted in each of the densely played video segments according to the length of each of the densely played video segments from long to short.
  • an embodiment of the present invention further provides a server, including: a transceiver and a processor, where
  • the transceiver is configured to count the playing intensity of each video frame in the video
  • the processor is configured to select at least one selected video frame from the video according to the playing density, wherein the playing density of the video frame is the number of times the video frame is played and the number of times the video is played. And combining the selected video frames to form the set of video frames.
  • the selecting, by the processor, the at least one selected video frame from the video comprises: screening a densely played video frame whose playback intensity is greater than a predetermined threshold; and filtering the filtered densely played video frame to form at least one densely played video segment;
  • the selected video frame is extracted from the at least one densely played video segment; wherein the consecutive densely played video frames constitute a densely played video segment.
  • the selecting, by the processor, the at least one selected video frame from the video further comprises: determining the number of the selected video frames extracted in each of the densely played video segments.
  • the processor determines the number of the selected video frames extracted in each of the densely played video segments, including: calculating an average playback intensity of all the video frames in each of the densely played video segments; calculating each of the densely played a proportion of a video segment in a video segment whose playback intensity is greater than or equal to the average playback density; and determining a number of the selected video frames extracted in each of the densely played video segments according to the ratio.
  • the processor determines, according to the ratio, the number of the selected video frames extracted in each of the densely-played video segments, including: gradually reducing the each intensive according to the ratio from large to small Plays the number of selected video frames extracted from the video clip.
  • the processor determines one of the selected video frames extracted in each of the densely played video segments The number includes: determining a length of each densely played video segment; and determining a number of the selected video frames extracted in each of the densely played video segments according to the length of each of the densely played video segments.
  • the determining, according to the length of each of the densely-played video segments, the number of the selected video frames extracted in each of the densely-played video segments includes: according to the length of each of the densely-played video segments From long to short, the number of selected video frames extracted in each of the densely played video segments is gradually reduced.
  • FIG. 11 is a schematic structural diagram of an application example of a server according to an embodiment of the present invention.
  • the terminal 500 includes: a processor 510, a memory 520, a transceiver 530, and a bus 540.
  • the processor 510, the memory 520, and the transceiver 530 are connected to each other through a bus 540.
  • the bus 540 may be an ISA bus, a PCI bus, or an EISA bus.
  • the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in Figure 5, but it does not mean that there is only one bus or one type of bus.
  • the memory 520 is configured to store a program.
  • the program can include program code, the program code including computer operating instructions.
  • Memory 520 may include high speed RAM memory and may also include non-volatile memory, such as at least one disk memory.
  • the transceiver 530 is configured to count the playing intensity of each video frame in the video.
  • the processor 510 executes the program code stored in the memory 520, for selecting at least one selected video frame from the video according to the play density, wherein the video frame is densely populated by the video. A ratio of the number of times the frame is played to the number of times the video is played, and the selected video frame is spliced to form the set of video frames.
  • the processor 510 selects at least one selected video frame from the video, including: screening a densely played video frame whose playback density is greater than a predetermined threshold; and filtering the filtered densely played video frames to form at least one dense play. a video segment; extracting a selected video frame from the at least one densely played video segment; wherein the consecutive densely played video frames constitute a densely played video segment.
  • the processor 510 selects at least one selected video frame from the video, and further includes: determining a number of the selected video frames extracted in each of the densely played video segments.
  • the processor 510 determines a number of the selected video frames extracted in each of the densely played video segments, including: calculating an average play density of all the video frames in each of the densely played video segments; Video in a densely played video clip with a play density greater than or equal to the average play density The proportion of frames; and determining the number of selected video frames extracted in each of the densely played video segments based on the ratio.
  • the processor 510 determines, according to the ratio, the number of the selected video frames extracted in each of the densely played video segments, including: gradually decreasing according to the ratio from large to small. The number of selected video frames extracted in each densely played video segment.
  • the processor 510 determines the number of the selected video frames extracted in each of the densely played video segments, including: determining a length of each densely played video segment; and determining, according to each of the densely played video segments Length, determining the number of selected video frames extracted in each of the densely played video segments.
  • the processor 510 determines, according to the length of each of the densely played video segments, the number of the selected video frames extracted in each of the densely played video segments, including: performing, according to each of the densely played The length of the video clip is from long to short, and the number of selected video frames extracted in each of the densely played video segments is gradually reduced.
  • the processor 510 is further configured to: when the number of actual video frames in each of the densely played video segments is less than the number of the selected selected video frames, extract the densely played video segment. All video frames in the frame, and the video frames are used as the number of the selected video frames; or the number of actual video frames in each of the densely played video segments is greater than or equal to the extracted When the number of selected video frames is selected, successive frames of each of the densely played video segments are randomly extracted, and the consecutive frames are used as the number of the selected selected video frames.
  • the processor 510 splicing the selected video frames to form the video frame set includes: grading the extracted selected video frames in a stepwise manner; wherein, the same densely played video segment The selected video frames are arranged in the order in which they are played in the video.
  • the processor 510 extracts the selected video frame from the at least one densely played video segment, including: extracting a predetermined number of video frames in each of the densely played video segments; or playing in each of the densely played video segments All video frames are extracted from the video clip.
  • the processor 510 extracts the selected video frame from the at least one densely played video segment, including: randomly extracting the selected video frame in each of the densely played video segments; or extracting the intensity of each densely played video segment Several video frames preceding the largest video frame are used as the selected video frame.
  • the processor 510 selects at least one selected video frame from the video, including: selecting a predetermined predetermined number of video frames of the maximum broadcast video frame as the selected video frame.
  • the processor 510 splicing the selected video frames to form the video frame set, including: sequentially splicing the selected video frames according to a play time sequence of the selected video frames in the video, Forming the set of video frames.
  • the processor may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or the like.
  • the computer storage medium may store a program, which may include some or all of the steps in various embodiments for generating a set of video frames provided by embodiments of the present invention.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).
  • the disclosed systems, devices, and methods may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • the functions, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium, including computer executed
  • the instructions when executed by a processor of a computer to execute the computer-executed instructions, perform the steps of any of the methods of generating a set of video frames described above.
  • the technical solution of the present invention which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) or a processor to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Databases & Information Systems (AREA)
  • Computer Security & Cryptography (AREA)
  • Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
  • Television Signal Processing For Recording (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本发明公开了一种产生视频帧集合的方法、设备及服务器。所述方法包括：统计视频中各个视频帧的播放密集程度，根据所述播放密集程度，从视频中选择至少一个"中选视频帧"，其中视频帧的播放密集程度是该视频帧被播放次数与视频被播放次数之比。然后，拼接"中选视频帧"以形成视频帧集合。本实施例通过视频中各个视频帧的密集播放程度来选择"中选视频帧"，并将由此选择的视频帧拼接在一起形成视频帧集合，能够将广大用户较多地认为精彩的视频内容拼接到视频帧集合中。

Description

产生视频帧集合的方法、设备及服务器
本发明要求于2014年10月31日提交中国专利局、申请号为201410610673.9、发明名称为“产生视频帧集合的方法和设备”的中国专利申请的优先权，其全部内容通过引用结合在本发明中。
技术领域
本发明涉及互联网领域，特别涉及产生视频帧集合的方法、设备及服务器。
背景技术
随着网络的发展，各种视频资源(如电视剧、电影、以及各种综艺节目)数量庞大，一个视频网站往往整合了上千上万个视频资源。
一般的综艺节目视频的播放时往往长达2个小时以上。若是电视剧，播放时长就更长了。
而对于用户来说，他们可能仅仅关心视频中的精彩部分，若需要看完视频的整体内容需要浪费不少时间。
为节约用户观看视频的时间，一些视频资源提供商提供精彩视频(精彩视频帧集合)，如某部电影的精彩预告、某一集电视剧的精彩预告等，以吸引用户的关注。视频资源提供商根据自己的认识和理解，从整个视频资源中抽取他们认为的精彩视频片段，进而合成这样的精彩视频。
然而，这种精彩视频是固定不变的，是在提供商的角度所认为的精彩视频，不能真实地体现用户的角度的精彩视频，存在用户错过观看多数用户所认为的精彩视频的问题。
发明内容
本发明所要解决的一个技术问题是，提供一种产生视频帧集合的方法、设备及服务器，其能够将原始视频(也可以称为“原视频”)中广大用户较多地认为精彩的视频内容拼接到视频帧集合中。
根据本发明的一个方面，提供了一种产生视频帧集合的方法，包括：统计视 频中各个视频帧的播放密集程度；根据所述播放密集程度，从视频中选择至少一个中选视频帧，其中所述视频帧的播放密集程度是该视频帧被播放次数与视频被播放次数之比；以及拼接中选视频帧以形成视频帧集合。
优选地，该方法还可以包括：记录中选视频帧在视频中的时间戳。
优选地，从视频中选择至少一个中选视频帧的步骤包括：筛选出播放密集程度大于预定阈值的密集播放视频帧，将筛选出的所述密集播放视频帧构成至少一个密集播放视频片段；从至少一个密集播放视频片段中抽取中选视频帧；其中，连续的所述密集播放视频帧构成一个密集播放视频片段。
优选地，从视频中选择至少一个中选视频帧的步骤还可以包括：确定要在每个密集播放视频片段中抽取的中选视频帧的个数。
优选地，确定要在每个密集播放视频片段中抽取的中选视频帧的个数的步骤包括：计算每个密集播放视频片段中所有视频帧的平均播放密集程度；计算每个密集播放视频片段中播放密集程度大于或等于平均播放密集程度的视频帧所占比例；以及根据比例，确定要在每个密集播放视频片段中抽取的中选视频帧的个数。
优选的，所述根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数的步骤包括：根据所述比例从大到小，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。其中在比例较大的密集播放视频片段中抽取的中选视频帧的个数大于或等于在比例较小的密集播放视频片段中抽取的中选视频帧的个数。
优选地，当密集播放视频片段中的实际视频帧个数小于所确定的要在该密集播放视频片段中抽取的中选视频帧的个数时，抽取密集播放视频片段中的所有视频帧。
优选地，确定要在每个密集播放视频片段中抽取的中选视频帧的个数的步骤包括：确定每个密集播放视频片段的长度，即每个密集播放视频片段中所包含的视频帧的个数；以及根据所述每个密集播放视频片段的长度，确定要在每个密集播放视频片段中抽取的中选视频帧的个数。
优选的，所述根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数的步骤包括：根据所述每个密集播放视 频片段的长度从长到短，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。其中在长度较长的密集播放视频片段中抽取的中选视频帧的个数大于或等于在长度较短的密集播放视频片段中抽取的中选视频帧的个数。
优选地，当所述每个密集播放视频片段中的实际视频帧个数小于所述抽取的所述中选视频帧的个数时，抽取所述密集播放视频片段中的所有视频帧，并将所述所有视频帧作为抽取的所述中选视频帧的个数；或者
当所述每个密集播放视频片段中的实际视频帧个数大于或等于所述抽取的所述中选视频帧的个数时，随机抽取所述每个密集播放视频片段的连续帧，并将所述连续帧作为抽取的所述中选视频帧的个数。
优选地，所述拼接所述中选视频帧以形成所述视频帧集合的步骤包括：将所述抽取的中选视频帧按照由多到少逐步排列；其中，同一个密集播放视频片段中的中选视频帧按其在所述视频中的播放时间顺序排列。
优选地，在从至少一个密集播放视频片段中抽取中选视频帧的步骤中，在每个密集播放视频片段中都抽取预定数量的视频帧；或者在每个密集播放视频片段中都抽取全部视频帧。
优选地，在从至少一个密集播放视频片段中抽取中选视频帧的步骤中，在所述每个密集播放视频片段中随机抽取中选视频帧；或者抽取各个密集播放视频片段中密集程度最大的的视频帧之前若干个视频帧作为中选视频帧。
优选地，在从视频中选择至少一个中选视频帧的步骤包括，选择播放密集程度最大视频帧的前预定数量个视频帧作为中选视频帧。
优选地，在拼接中选视频帧以形成视频帧集合的步骤包括，根据中选视频帧在视频中的播放时间顺序，依次拼接所选择的视频帧，以形成所述视频帧集合。
根据本发明的另一个方面，提供了一种产生视频帧集合的设备，包括：统计装置，用于统计视频中各个视频帧的播放密集程度；视频帧选择装置，用于根据视频中各个视频帧的播放密集程度从视频中选择至少一个中选视频帧，其中视频帧的播放密集程度是该视频帧被播放次数与视频被播放次数之比；以及拼接装置，用于拼接中选视频帧以形成视频帧集合。
优选地，该设备还可以包括：统计装置，用于统计播放密集程度；和/或时间戳记录装置，用于记录中选视频帧在视频中的时间戳。
优选地，视频帧选择装置可以包括：视频帧筛选装置，用于筛选出播放密集程度大于预定阈值的密集播放视频帧，将筛选出的所述密集播放视频帧构成至少一个密集播放视频片段；其中连续的密集播放视频帧构成密集播放视频片段；以及视频帧抽取装置，用于从至少一个密集播放视频片段中抽取中选视频帧。
优选地，视频帧选择装置还可以包括：帧数确定装置，用于确定要在每个密集播放视频片段中抽取的中选视频帧的个数。
优选地，帧数确定装置可以包括：平均播放密集程度计算装置，用于计算每个密集播放视频片段中所有视频帧的平均播放密集程度；比例计算装置，用于计算每个密集播放视频片段中播放密集程度大于或等于平均播放密集程度的视频帧所占比例；以及第一帧数确定装置，用于根据比例，确定要在每个密集播放视频片段中抽取的中选视频帧的个数。
优选地，第一帧数确定装置具体用于根据所述比例从大到小，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。即在比例较大的密集播放视频片段中抽取较多的中选视频帧，而在比例较小的密集播放视频片段中抽取较少的中选视频帧。
优选地，帧数确定装置可以包括：视频片段长度确定装置，用于确定每个密集播放视频片段的长度，即每个密集播放视频片段中所包含的视频帧的个数；以及第二帧数确定装置，用于根据密集播放视频片段的长度，确定要在每个密集播放视频片段中抽取的中选视频帧的个数。
优选地，所述第二帧数确定装置，具体用于根据所述每个密集播放视频片段的长度从长到短，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。中在长度较长的密集播放视频片段中抽取较多的中选视频帧，而在长度较短的密集播放视频片段中抽取较少的中选视频帧。
根据本发明的另一个方面，提供了一种服务器，包括：
收发器，用于统计视频中各个视频帧的播放密集程度；
处理器，用于根据所述播放密集程度，从所述视频中选择至少一个中选视频帧，其中所述视频帧的播放密集程度是所述视频帧被播放次数与所述视频被播放次数之比；以及拼接所述中选视频帧以形成所述视频帧集合。
优选地，所述处理器从视频中选择至少一个中选视频帧包括：筛选出播放密 集程度大于预定阈值的密集播放视频帧；将筛选出的所述密集播放视频帧构成至少一个密集播放视频片段；从所述至少一个密集播放视频片段中抽取中选视频帧；其中，连续的所述密集播放视频帧构成一个密集播放视频片段。
优选地，所述处理器从视频中选择至少一个中选视频帧还包括：确定在每个密集播放视频片段中抽取的中选视频帧的个数。
优选地，所述处理器确定在每个密集播放视频片段中抽取的中选视频帧的个数，包括：计算每个密集播放视频片段中所有视频帧的平均播放密集程度；计算所述每个密集播放视频片段中播放密集程度大于或等于所述平均播放密集程度的视频帧所占比例；以及根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
优选地，所述处理器根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数，包括：根据所述比例从大到小，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
优选地，所述处理器确定在每个密集播放视频片段中抽取的中选视频帧的个数，包括：确定每个密集播放视频片段的长度；以及根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
优选地，所述处理器根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数包括：根据所述每个密集播放视频片段的长度从长到短，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
本发明实施例还提供一种计算机可读存储介质，包括计算机执行指令，以供计算机的处理器执行所述计算机执行指令时，所述计算机执行上述产生视频帧集合的方法的全部或部分步骤。
本发明通过根据视频中个视频帧的密集播放程度来选择中选视频帧，并将由此选择的视频帧拼接在一起形成视频帧集合，能够将广大用户较多地认为精彩的视频内容拼接到视频帧集合中。
附图说明
图1是根据本发明的第一实施例的产生视频帧集合的方法的流程图。
图2是根据本发明的第二实施例的产生视频帧集合的方法的流程图。
图3是图1和图2中的步骤S200的一种实现方式的流程图。
图4是图3中的步骤S220的一种实现方式的流程图。
图5是图3中的步骤S220的另一种实现方式的流程图。
图6是根据本发明的第三实施例的产生视频帧集合的设备的方框图。
图7是根据本发明的第四实施例的产生视频帧集合的设备的方框图。
图8是图6和图7中的视频帧选择装置200可以包括的子装置的方框图。
图9是图8中的帧数确定装置220的一种实现方式的方框图。
图10是图8中的帧数确定装置220的另一种实现方式的方框图。
图11是本发明实施例提供的一种服务器的结构示意图。
具体实施方式
通过本方案形成的精彩视频，可以反映用户角度的精彩视频，形成的精彩视频可以根据用户播放的情况实时变化，可以将用户认为最精彩的视频展现给其他用户，使得用户能够观看多数用户所认为的精彩视频。
下面参考附图描述根据本发明的产生视频帧集合的方法和设备。
图1是根据本发明的第一实施例的产生视频帧集合的方法的流程图。
首先，在步骤S100，统计视频中各个视频帧的播放密集程度。其中，播放密集程度是指播放该视频帧的次数与观看该视频次数之比。如：播放时间戳为1:00:00处的视频帧的次数为80次，播放该视频的次数为100次，则该视频帧的播放密集程度为80％。
其具体的统计方式有多种，本实施例以一种方式为例，具体包括：
1)通过视频播放服务器记录某个视频的播放信息；
也就是说，当有用户请求播放视频(即发送视频播放请求)时，视频播放服务器记录用户观看了哪些视频帧，记录这个视频被播放一次，这些被观看的视频帧被播放一次。
其中，该播放信息可以包括：各视频帧的播放次数、播放该视频的次数(可 认为是点击该视频连接进行播放的次数)和/或各视频帧的时间戳等等。
2)根据记录的所述播放信息计算各视频帧的播放密集程度：
在统计一定数量的视频播放请求之后，根据统计到的播放信息计算每个视频帧的播放密集程度，从而得到该视频帧被播放的次数与该视频被播放的次数之比。
其中，播放密集程度可以由视频提供商自行统计，也可以委托其它机构统计，或者从其它数据库直接获取。
但是，作为视频提供商，往往掌握第一手的播放记录，能够更加准确并迅速地实现这样的统计。播放密集程度可以随着更多用户的使用不断更新。
其次，在步骤S200，根据视频中各个视频帧的播放密集程度，从视频中选择至少一个“中选视频帧”。在这里，“中选视频帧”是指从原视频中挑选出来，要放入视频帧集合中的视频帧。
在这里，视频帧的“播放密集程度”是该视频帧被播放的次数与其所在的视频被播放的次数之比。
当视频被播放时，用户可能指令快进，可能跳跃，也可能提前结束，并不是每个视频帧都一定会被播放。因此，有些视频帧被播放的次数多，有些视频帧被播放的次数少。通过统计大量用户的行为，可以判断哪些视频帧是用户更加感兴趣的。因此，通过根据各个视频帧的播放密集程度来选择“中选视频帧”，更有可能挑选出用户更感兴趣的内容。
视频帧的密集播放程度可以是由视频提供商自行统计的，也可以是委托专门的调查公司或其它机构统计的，也可以是从其它数据库中直接获取的。
最后，在挑选好“中选视频帧”之后，在步骤S400，拼接这些“中选视频帧”，从而形成视频帧集合。
在拼接“中选视频帧”时，可以按照这些“中选视频帧”在原视频中的播放顺序，依次拼接。
也可以根据播放密集程度，从高到低按顺序拼接这些“中选视频帧”。这样用户有可能先看到更多用户关注的内容。
通过在选择“中选视频帧”是考虑视频帧的密集播放程度，可以挑选出有更 多用户观看过的视频帧，从而更有可能挑选出广大用户更加喜欢的内容。
图2是根据本发明的第二实施例的产生视频帧集合的方法的流程图。
图2所示的方法比图1所示的方法多了一个步骤，步骤S300，其他几个步骤详见图1，在此不再赘述。
在图2所示的方法的步骤S100中，先统计视频中各个视频帧的播放密集程度，其统计的过程详见上述，在此不再赘述。
其中，播放密集程度可以随着更多用户的使用不断更新。
本方法可以在统计了足够多的用户的使用记录之后执行后续的步骤。也可以在统计了一定数量的用户使用记录之后执行后续的步骤，以选择“中选视频帧”，并形成视频帧集合；然后在进一步统计了一定数量的用户使用记录之后，使用新的播放密集程度，重新选择“中选视频帧”，重新形成新的视频帧集合，以便不断优化视频帧集合的内容。
然后和上面参考图1描述的一样，在步骤S200，根据统计视频中各个视频帧的播放密集程度，从视频中选择至少一个“中选视频帧”。
然后，在步骤S300，可以进一步记录各个“中选视频帧”在原视频中的时间戳。这样可以在播放视频帧集合时，容易的找到正在播放的视频帧所对应的原视频中的位置。当用户在观看视频帧集合时，看到感兴趣的内容时，可以从感兴趣的地方方便地进入原视频相应位置开始正常观看。
可以预先记录好所有视频帧的时间戳，每个视频帧与其时间戳对应。这样，可以不必专门再记录“中选视频帧”的时间戳。只要能够观看视频帧集合时，指定某个视频帧，能够确定其在原视频中的时间戳，从而能够确定其在原视频中的视频位置，就可以。
当然，也可以在步骤S200中选择出“中选视频帧”后，专门对应地记录“中选视频帧”的时间戳和“中选视频帧”的识别代码。
在步骤S400，与参考图1描述的一样，拼接这些“中选视频帧”，从而形成视频帧集合。
也就是说，在挑选好“中选视频帧”之后，在步骤S400，拼接这些“中选视频帧”，从而形成视频帧集合。
在拼接“中选视频帧”时，可以按照这些“中选视频帧”在原视频中的播放顺序，依次拼接。
也可以根据播放密集程度，从高到低按顺序拼接这些“中选视频帧”。这样用户有可能先看到更多用户关注的内容。
本发明实施例中，通过在选择“中选视频帧”是考虑视频帧的密集播放程度，可以挑选出有更多用户观看过的视频帧，从而更有可能挑选出广大用户更加喜欢的内容。
需要说明的，步骤S400与步骤S300的执行顺序可以颠倒，也可以同时执行。
下面，对图1和图2中的步骤S200做进一步的描述。
在步骤S200中，可以对各视频帧的播放密集程度进行排序。然后，直接选择播放密集程度最大的前预定数量个视频帧作为“中选视频帧”。预定数量可以根据需要来定。
这样，“中选视频帧”是用户播放次数最多的若干个视频帧，比较容易体现用户关注哪些视频帧。
然而，也有可能播放次数最多的预定数量个视频帧是在连续的视频片段中，这样，视频帧集合就有可能不能反映视频中更多的内容。
另外，还有可能，有些视频帧只是用户在快进时播放到的，虽然被统计为被播放次数较多，但其实前后的视频帧都没有被播放过，实际上并不是用户真正关注的视频帧。
为了克服上述问题，这里进一步提出“密集播放视频片段”的概念。
下面参考图3至图5进一步描述借助密集播放视频片段来执行上面参考图1和图2描述的步骤S200的方法。
图3是图1和图2中的步骤S200的一种实现方式的流程图。
首先，在步骤S210，筛选出视频中播放密集程度大于预定阈值的视频帧，在此称为“密集播放视频帧”。
这里的预定阈值可以是对所有视频预先指定的相同的值，例如60％。
预定阈值也可以针对不同的视频而有所不同。
另外，这个预定阈值还可以根据视频中各视频帧的实际播放情况来决定。例如可以设定为最高播放密集程度的一定比例，例如最高播放密集程度的一半、三分之二等。
这样筛选出的密集播放视频帧中会有一些连续的密集播放视频帧，这些连续的密集播放视频帧构成一个个视频片段，在此称为“密集播放视频片段”。多个视频帧是否连续，可以根据其时间戳来判断。当一个视频帧不与其它视频帧相连时，也可以视为，该视频帧单独构成了一个视频片段，只是这个视频片段只有一个视频帧而已。
这样，所筛选出的多个密集播放视频帧将会构成至少一个密集播放视频片段。
然后，在步骤S220，确定要在每个密集播放视频片段中抽取的“中选视频帧”的个数。该步骤为可选步骤。
然后，在步骤S230，从上述至少一个密集播放视频片段中抽取“中选视频帧”。
在步骤S230中，在密集播放视频片段中抽取“中选视频帧”时，可以从密集播放视频片段中随机地抽取“中选视频帧”，也可以抽取各个密集播放视频片段中密集程度最大的前若干个视频帧作为“中选视频帧”。
另外，在步骤S230中，在每个密集播放视频片段中抽取的“中选视频帧”可以是连续的，也可以是不连续的。如果抽取连续的“中选视频帧”，则可以向用户呈现较为连续的小视频片段。
具体抽取数量可以通过上述步骤S220确定，也可以不通过上述步骤S220确定。
图3示出了包括需要确定要在每个密集播放视频片段中抽取的“中选视频帧”的个数的步骤的情况。
然而，也可以不经过步骤S220，而是预先规定在步骤S230中要在每个视频帧中抽取多少个“中选视频帧”。
例如，可以规定在每个密集播放视频片段中都抽取相同数量(预定数量)的视频帧，例如，在每个密集播放视频片段中都抽取5个视频帧。
或者，例如，在步骤S210中所设定的预定阈值较高时，有可能所筛选出的密集播放视频帧较少，从而也可以在每个密集播放视频片段中都抽取全部视频帧。
这样，就不需要上述步骤S220，而是在步骤S230中直接抽取规定数量的“中选视频帧”即可。
然而，通过步骤S220，根据实际视频播放统计情况来确定要在每个密集播放视频片段中抽取的视频帧数量，能够更好地反应用户的视角。
下面，参考图4和图5描述两种确定要抽取的视频帧数量的方式。
图4是图3中的步骤S220的一种实现方式的流程图。
在图4所示实现方式中，根据密集播放视频片段中个视频帧的播放密集程度分布情况来决定要在该密集播放视频片段中抽取多少个视频帧。
在步骤S222，计算每个密集播放视频片段中所有视频帧的平均播放密集程度。
然后，在步骤S223，计算每个密集播放视频片段中播放密集程度大于或等于平均播放密集程度的视频帧所占比例。
当这个比例较小时，说明这个视频片段中有可能只有个别视频因为快进等一些偶然因素而被播放较多次。因此，可以认为，这个视频片段的重要性要低一些，可以从这个视频片段中抽取较少的视频帧。
相反，当这个比例较大时，说明这个视频片段中的视频帧可能较为普遍地被关注。因此，可以认为这个视频片段的重要性要高一些，可以从这个视频片段中抽取较多的视频帧。
然后，可以在步骤S224中，根据步骤S223中计算的比例，来确定要在每个密集播放视频片段中抽取的“中选视频帧”的个数。
具体说来，可以在上述比例较大的密集播放视频片段中抽取较多的“中选视频帧”，而在上述比例较小的密集播放视频片段中抽取较少的“中选视频帧”。
可以按照上述比例从高到低(或从大到小)的顺序逐步减少所抽取的“中选视频帧”的个数。
例如，视频片段1的上述比例最高，为90％，视频片段2的上述比例次高， 为80％，……则，从视频片段1中抽取5帧，从视频片段2中抽取4帧……。
如上所述，在一个密集播放视频片段中，可以抽取连续的帧，也可以抽取不连续的帧；可以随机抽取帧，也可以抽取播放密集程度最大的几个帧。
在使用图4所示方式确定要抽取的帧数时，可能存在这样的情况，即某个密集播放视频片段中的实际视频帧个数小于按照上述比例所确定的要在该密集播放视频片段中抽取的“中选视频帧”的个数。此时，可以抽取该密集播放视频片段中的所有视频帧。并将所有视频帧作为抽取的“中选视频帧”的个数。
也就是说，当所述每个密集播放视频片段中的实际视频帧个数小于所述抽取的所述中选视频帧的个数时，抽取所述密集播放视频片段中的所有视频帧，并将所述所有视频帧作为抽取的所述中选视频帧的个数；或者
当所述每个密集播放视频片段中的实际视频帧个数大于或等于所述抽取的所述中选视频帧的个数时，随机抽取所述每个密集播放视频片段的连续帧，并将所述连续帧作为抽取的所述中选视频帧的个数。
图5是图3中的步骤S220的另一种实现方式的流程图。
在图5所示实现方式中，根据密集播放视频片段的长度(即视频帧个数)来决定要在该密集播放视频片段中抽取多少个视频帧。
首先，在步骤S226，确定每个密集播放视频片段的长度。在这里，密集播放视频片段的长度可以用密集播放视频片段中所包含的视频帧的个数来表示。
然后，在步骤S228，根据密集播放视频片段的长度，确定要在每个密集播放视频片段中抽取的“中选视频帧”的个数。
可以在长度较长的密集播放视频片段中抽取较多的“中选视频帧”，而在长度较短的密集播放视频片段中抽取较少的“中选视频帧”。
可以按照密集播放视频片段长度从大到小(或从长到短)的顺序逐步减少所抽取的“中选视频帧”的个数。
例如，从最长的密集播放视频片段中抽取5帧，从第二长的密集播放视频片段中抽取4帧，……
同样地，在一个密集播放视频片段中，可以抽取连续的帧，也可以抽取不连续的帧；可以随机抽取帧，也可以抽取播放密集程度最大的几个帧。
上文中参考图4和5描述可以用于在步骤S220中确定要在每个密集播放视频片段中抽取的“中选视频帧”的个数的两种方式。
在通过步骤S220确定了要抽取的“中选视频帧”的个数的情况下，可以认为，从中抽选了较多“中选视频帧”的视频片段更加收到用户的关注。
因此，可以将所抽取“中选视频帧”较多的密集播放视频片段中抽取的“中选视频帧”排列在前，而将所抽选“中选视频帧”较少的密集播放视频片段中抽取的“中选视频帧”排列在后。从而将可能更受用户关注的内容放在视频帧集合的前面。
这样，虽然用户在观看视频帧集合时，并不是按照在原视频中的顺序观看各个“中选视频帧”的视频内容，但是有可能能够让用户更早观看到更受广大用户关注的“中选视频帧”的视频内容。
另一方面，所述拼接所述中选视频帧以形成所述视频帧集合，包括：将所述抽取的中选视频帧按照由多到少逐步排列；其中，
同一个密集播放视频片段中的“中选视频帧”可以按其在视频中的播放时间顺序排列，这样可以保证一个视频片段内视频内容的时间先后关系。
可选的，在另一实施例中，所述从至少一个密集播放视频片段中抽取中选视频帧，包括：在所述每个密集播放视频片段中都抽取预定数量的视频帧；或者在所述每个密集播放视频片段中都抽取全部视频帧。
可选的，在另一实施例中，所述从至少一个密集播放视频片段中抽取中选视频帧，包括：在所述每个密集播放视频片段中随机抽取中选视频帧；或者抽取各个密集播放视频片段中密集程度最大的视频帧之前的若干个视频帧作为中选视频帧。
下面，参考图6至图10，详细描述根据本发明的产生视频帧集合的设备，该设备可以用来执行上面参考图1至图5描述的产生视频帧集合的方法。
在下文中，主要描述了该设备的主要结构，其具体操作细节完全可以与上文中描述的一样。为了避免重复，下文中省略了对一些细节的描述。
图6是根据本发明的第三实施例的产生视频帧集合的设备的方框图。
图6所示设备包括统计装置100、视频帧选择装置200和拼接装置400。
统计装置100统计视频中各个视频帧的播放密集程度；具体包括：信息记录 装置，用于记录各个视频的播放信息；计算装置，用于根据所述播放信息计算一个视频中各视频帧的播放密集程度。其具体的实现过程详见上述，在此不再赘述。
视频帧选择装置200根据视频中各个视频帧的播放密集程度从视频中选择至少一个“中选视频帧”。如上所述，视频帧的播放密集程度是该视频帧被播放次数与视频被播放次数之比。视频帧选择装置200可以按与上面参考图1至图5描述的内容相同的方式来选择“中选视频帧”。
拼接装置400拼接视频帧选择装置200所选择的“中选视频帧”以形成视频帧集合。拼接装置400可以按与上面参考图1至图5描述的内容相同的方式来拼接“中选视频帧”。
图7是根据本发明的第四实施例的产生视频帧集合的设备的方框图。
与图6相比，图7所示设备增加了时间戳记录装置300。
时间戳记录装置300记录“中选视频帧”在视频中的时间戳。
图8是图6和图7中的视频帧选择装置200可以包括的子装置的方框图。
如图8所示，视频帧选择装置200可以包括视频帧筛选装置210、帧数确定装置220和视频帧抽取装置230。当然，也可以只包括：视频帧筛选装置200和视频帧抽取装置230(图中未示)，即为了限定精彩视频的大小，需要限定所抽取的“中选视频帧”的个数。
视频帧筛选装置210筛选出播放密集程度大于预定阈值的密集播放视频帧。其中连续的密集播放视频帧构成密集播放视频片段，所筛选出的密集播放视频帧构成至少一个密集播放视频片段。
帧数确定装置220确定要在每个密集播放视频片段中抽取的“中选视频帧”的个数。
视频帧抽取装置230从至少一个密集播放视频片段中抽取“中选视频帧”。
根据上面对根据本发明的方法的描述一样的理由，帧数确定装置220不是必须要有的。可以预先规定要在每个密集播放视频片段中抽取的帧数，例如都抽取预定数量的视频帧，或都抽取全部视频帧。
在使用帧数确定装置220的情况下，可以通过下述两种方式实现帧数确定装置220。
图9是图8中的帧数确定装置220的一种实现方式的方框图。
如图9所示，帧数确定装置220可以包括平均播放密集程度计算装置222、比例计算装置223和第一帧数确定装置224。
平均播放密集程度计算装置222计算每个密集播放视频片段中所有视频帧的平均播放密集程度。
比例计算装置223计算每个密集播放视频片段中播放密集程度大于平均播放密集程度的视频帧所占比例。
第一帧数确定装置224根据上述比例，确定要在每个密集播放视频片段中抽取的“中选视频帧”的个数。
可选的，所述第一帧数确定装置，具体用于根据所述比例从大到小，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。也就是说，在上述比例较大的密集播放视频片段中抽取较多的“中选视频帧”，而在上述比例较小的密集播放视频片段中抽取较少的“中选视频帧”。
当密集播放视频片段中的实际视频帧个数小于所确定的要在该密集播放视频片段中抽取的“中选视频帧”的个数时，抽取密集播放视频片段中的所有视频帧，并将所述所有视频帧作为抽取的所述中选视频帧的个数；或者当所述每个密集播放视频片段中的实际视频帧个数大于或等于所述抽取的所述中选视频帧的个数时，随机抽取所述每个。
图10是图8中的帧数确定装置220的另一种实现方式的方框图。
如图10所示，帧数确定装置220可以包括视频片段长度确定装置226和第二帧数确定装置228。
视频片段长度确定装置226确定每个密集播放视频片段的长度，即每个密集播放视频片段中所包含的视频帧的个数。
第二帧数确定装置228根据密集播放视频片段的长度，确定要在每个密集播放视频片段中抽取的“中选视频帧”的个数。
第二帧数确定装置228具体根据所述每个密集播放视频片段的长度从长到短，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
其中在长度较长的密集播放视频片段中抽取较多的“中选视频帧”，而在长 度较短的密集播放视频片段中抽取较少的“中选视频帧”。也就是说，当所述每个密集播放视频片段中的实际视频帧个数小于所述抽取的所述中选视频帧的个数时，抽取所述密集播放视频片段中的所有视频帧，并将所述所有视频帧作为抽取的所述中选视频帧的个数；或者当所述每个密集播放视频片段中的实际视频帧个数大于或等于所述抽取的所述中选视频帧的个数时，随机抽取所述每个密集播放视频片段的连续帧，并将所述连续帧作为抽取的所述中选视频帧的个数。
至此，已详细描述了根据本发明的产生视频帧集合的方法和设备的具体实施例。然而本领域技术人员应该明白，本发明不限于这里描述的各种细节，而是可以做出适当的修改。本发明的保护范围由所附权利要求书限定。
相应，本发明实施例还提供一种服务器，包括：收发器和处理器，其中，
所述收发器，用于统计视频中各个视频帧的播放密集程度；
所述处理器，用于根据所述播放密集程度，从所述视频中选择至少一个中选视频帧，其中所述视频帧的播放密集程度是所述视频帧被播放次数与所述视频被播放次数之比；以及拼接所述中选视频帧以形成所述视频帧集合。
其中，所述处理器从视频中选择至少一个中选视频帧包括：筛选出播放密集程度大于预定阈值的密集播放视频帧；将筛选出的所述密集播放视频帧构成至少一个密集播放视频片段；从所述至少一个密集播放视频片段中抽取中选视频帧；其中，连续的所述密集播放视频帧构成一个密集播放视频片段。
其中，所述处理器从视频中选择至少一个中选视频帧还包括：确定在每个密集播放视频片段中抽取的中选视频帧的个数。
其中，所述处理器确定在每个密集播放视频片段中抽取的中选视频帧的个数，包括：计算每个密集播放视频片段中所有视频帧的平均播放密集程度；计算所述每个密集播放视频片段中播放密集程度大于或等于所述平均播放密集程度的视频帧所占比例；以及根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
其中，所述处理器根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数，包括：根据所述比例从大到小，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
其中，所述处理器确定在每个密集播放视频片段中抽取的中选视频帧的个 数，包括：确定每个密集播放视频片段的长度；以及根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
其中，所述处理器根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数包括：根据所述每个密集播放视频片段的长度从长到短，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
参见图11，为本发明实施例提供的一种服务器的应用实例的结构示意图，该终端500包括：处理器510、存储器520、收发器530和总线540；
处理器510、存储器520、收发器530通过总线540相互连接；总线540可以是ISA总线、PCI总线或EISA总线等。所述总线可以分为地址总线、数据总线、控制总线等。为便于表示，图5中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
存储器520，用于存放程序。具体地，程序可以包括程序代码，所述程序代码包括计算机操作指令。存储器520可能包含高速RAM存储器，也可能还包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。
所述收发器530，用于统计视频中各个视频帧的播放密集程度；
所述处理器510执行存储器520中存储的所述程序代码，用于根据所述播放密集程度，从所述视频中选择至少一个中选视频帧，其中所述视频帧的播放密集程度是所述视频帧被播放次数与所述视频被播放次数之比，拼接所述中选视频帧以形成所述视频帧集合。
可选地，所述处理器510从视频中选择至少一个中选视频帧，包括：筛选出播放密集程度大于预定阈值的密集播放视频帧；将筛选出的所述密集播放视频帧构成至少一个密集播放视频片段；从所述至少一个密集播放视频片段中抽取中选视频帧；其中，连续的所述密集播放视频帧构成一个密集播放视频片段。
可选地，所述处理器510从视频中选择至少一个中选视频帧，还包括：确定在每个密集播放视频片段中抽取的中选视频帧的个数。
可选地，所述处理器510确定在每个密集播放视频片段中抽取的中选视频帧的个数，包括：计算每个密集播放视频片段中所有视频帧的平均播放密集程度；计算所述每个密集播放视频片段中播放密集程度大于或等于所述平均播放密集程度的视频 帧所占比例；以及根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
可选地，所述处理器510根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数，包括：根据所述比例从大到小，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
可选地，所述处理器510确定在每个密集播放视频片段中抽取的中选视频帧的个数，包括：确定每个密集播放视频片段的长度；以及根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
可选地，所述处理器510根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数，包括：根据所述每个密集播放视频片段的长度从长到短，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
可选地，所述处理器510还用于在所述每个密集播放视频片段中的实际视频帧个数小于所述抽取的所述中选视频帧的个数时，抽取所述密集播放视频片段中的所有视频帧，并将所述所有视频帧作为抽取的所述中选视频帧的个数；或者在所述每个密集播放视频片段中的实际视频帧个数大于或等于所述抽取的所述中选视频帧的个数时，随机抽取所述每个密集播放视频片段的连续帧，并将所述连续帧作为抽取的所述中选视频帧的个数。
可选地，所述处理器510拼接所述中选视频帧以形成所述视频帧集合，包括：将所述抽取的中选视频帧按照由多到少逐步排列；其中，同一个密集播放视频片段中的中选视频帧按其在所述视频中的播放时间顺序排列。
可选地，所述处理器510从至少一个密集播放视频片段中抽取中选视频帧，包括：在所述每个密集播放视频片段中都抽取预定数量的视频帧；或者在所述每个密集播放视频片段中都抽取全部视频帧。
可选地，所述处理器510从至少一个密集播放视频片段中抽取中选视频帧，包括：在所述每个密集播放视频片段中随机抽取中选视频帧；或者抽取各个密集播放视频片段中密集程度最大的视频帧之前的若干个视频帧作为中选视频帧。
可选地，所述处理器510从视频中选择至少一个中选视频帧，包括：选择所述播放密集程度最大视频帧的前预定数量个视频帧作为所述中选视频帧。
可选地，所述处理器510拼接所述中选视频帧以形成所述视频帧集合，包括：根据所述中选视频帧在所述视频中的播放时间顺序，依次拼接所选择的视频帧，以形成所述视频帧集合。
所述服务器中的收发器和处理器的功能和作用的实现过程详见上述方法实施例中对应部分的实现过程，在此不再赘述。
具体实现中，上述处理器可以是中央处理器(central processing unit,CPU)、专用集成电路(applicatI/On-specific integrated circuit,ASIC)等。计算机存储介质可存储有程序，该程序执行时可包括本发明实施例提供的产生视频帧集合的各实施例中的部分或全部步骤。所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)等。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的装置和服务器的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中,该计算机可读取存储介质包括计算机执行 指令，以供计算机的处理器执行所述计算机执行指令时，所述计算机执行上述任一产生视频帧集合的方法的步骤。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器(processor)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应所述以权利要求的保护范围为准。

Claims (28)

  1. 一种产生视频帧集合的方法，包括：
    统计视频中各个视频帧的播放密集程度；
    根据所述播放密集程度，从所述视频中选择至少一个中选视频帧，其中所述视频帧的播放密集程度是所述视频帧被播放次数与所述视频被播放次数之比；以及
    拼接所述中选视频帧以形成所述视频帧集合。
  2. 根据权利要求1所述的方法，其中，所述从视频中选择至少一个中选视频帧的步骤包括：
    筛选出播放密集程度大于预定阈值的密集播放视频帧；
    将筛选出的所述密集播放视频帧构成至少一个密集播放视频片段；
    从所述至少一个密集播放视频片段中抽取中选视频帧；
    其中，连续的所述密集播放视频帧构成一个密集播放视频片段。
  3. 根据权利要求2所述的方法，其中，所述从视频中选择至少一个中选视频帧的步骤还包括：
    确定在每个密集播放视频片段中抽取的中选视频帧的个数。
  4. 根据权利要求3所述的方法，其中，所述确定在每个密集播放视频片段中抽取的中选视频帧的个数的步骤包括：
    计算每个密集播放视频片段中所有视频帧的平均播放密集程度；
    计算所述每个密集播放视频片段中播放密集程度大于或等于所述平均播放密集程度的视频帧所占比例；以及
    根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
  5. 根据权利要求4所述的方法，其中，所述根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数的步骤包括：
    根据所述比例从大到小，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
  6. 根据权利要求3所述的方法，其中，所述确定在每个密集播放视频片段中抽取的中选视频帧的个数的步骤包括：
    确定每个密集播放视频片段的长度；以及
    根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段 中抽取的中选视频帧的个数。
  7. 根据权利要求6所述的方法，其中，所述根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数的步骤包括：
    根据所述每个密集播放视频片段的长度从长到短，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
  8. 根据权利要求5或7所述的方法，其中，还包括：
    当所述每个密集播放视频片段中的实际视频帧个数小于所述抽取的所述中选视频帧的个数时，抽取所述密集播放视频片段中的所有视频帧，并将所述所有视频帧作为抽取的所述中选视频帧的个数；或者
    当所述每个密集播放视频片段中的实际视频帧个数大于或等于所述抽取的所述中选视频帧的个数时，随机抽取所述每个密集播放视频片段的连续帧，并将所述连续帧作为抽取的所述中选视频帧的个数。
  9. 根据权利要求3至6中任何一项所述的方法，其中，所述拼接所述中选视频帧以形成所述视频帧集合的步骤包括：
    将所述抽取的中选视频帧按照由多到少逐步排列；其中，
    同一个密集播放视频片段中的中选视频帧按其在所述视频中的播放时间顺序排列。
  10. 根据权利要求2所述的方法，其中，所述从至少一个密集播放视频片段中抽取中选视频帧的步骤包括：
    在所述每个密集播放视频片段中都抽取预定数量的视频帧；或者
    在所述每个密集播放视频片段中都抽取全部视频帧。
  11. 根据权利要求2所述的方法，其中，所述从至少一个密集播放视频片段中抽取中选视频帧的步骤包括：
    在所述每个密集播放视频片段中随机抽取中选视频帧；或者
    抽取各个密集播放视频片段中密集程度最大的视频帧之前的若干个视频帧作为中选视频帧。
  12. 根据权利要求1所述的方法，其中，所述从视频中选择至少一个中选视频帧的步骤包括：
    选择所述播放密集程度最大视频帧的前预定数量个视频帧作为所述中选视频帧。
  13. 根据权利要求1至6任一项或10至12中任何一项所述的方法，其中， 所述拼接所述中选视频帧以形成所述视频帧集合的步骤包括：
    根据所述中选视频帧在所述视频中的播放时间顺序，依次拼接所选择的视频帧，以形成所述视频帧集合。
  14. 一种产生视频帧集合的设备，包括：
    统计装置，用于统计视频中各个视频帧的播放密集程度；
    视频帧选择装置，用于根据所述播放密集程度从所述视频中选择至少一个中选视频帧，其中所述视频帧的播放密集程度是所述视频帧被播放次数与所述视频被播放次数之比；以及
    拼接装置，用于拼接所述中选视频帧以形成所述视频帧集合。
  15. 根据权利要求14所述的设备，其中，所述视频帧选择装置包括：
    视频帧筛选装置，用于筛选出播放密集程度大于预定阈值的密集播放视频帧，将筛选出的所述密集播放视频帧构成至少一个密集播放视频片段；其中连续的所述密集播放视频帧构成一个密集播放视频片段；以及
    视频帧抽取装置，用于从所述至少一个密集播放视频片段中抽取中选视频帧。
  16. 根据权利要求15所述的设备，其中，所述视频帧选择装置还包括：
    帧数确定装置，用于确定在每个密集播放视频片段中抽取的中选视频帧的个数。
  17. 根据权利要求16所述的设备，其中，所述帧数确定装置包括：
    平均播放密集程度计算装置，用于计算每个密集播放视频片段中所有视频帧的平均播放密集程度；
    比例计算装置，用于计算所述每个密集播放视频片段中播放密集程度大于或等于所述平均播放密集程度的视频帧所占比例；以及
    第一帧数确定装置，用于根据所述比例，确定在每个密集播放视频片段中抽取的中选视频帧的个数。
  18. 根据权利要求17所述的设备，其中，所述第一帧数确定装置，具体用于根据所述比例从大到小，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
  19. 根据权利要求16所述的设备，其中，所述帧数确定装置包括：
    视频片段长度确定装置，用于确定每个密集播放视频片段的长度；以及
    第二帧数确定装置，用于根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
  20. 根据权利要求19所述的设备，其中，所述第二帧数确定装置，具体用于根据所述每个密集播放视频片段的长度从长到短，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
  21. 一种服务器，包括：
    收发器，用于统计视频中各个视频帧的播放密集程度；
    处理器，用于根据所述播放密集程度，从所述视频中选择至少一个中选视频帧，其中所述视频帧的播放密集程度是所述视频帧被播放次数与所述视频被播放次数之比；以及拼接所述中选视频帧以形成所述视频帧集合。
  22. 根据权利要求21所述的服务器，其中，所述处理器从视频中选择至少一个中选视频帧包括：筛选出播放密集程度大于预定阈值的密集播放视频帧；将筛选出的所述密集播放视频帧构成至少一个密集播放视频片段；从所述至少一个密集播放视频片段中抽取中选视频帧；其中，连续的所述密集播放视频帧构成一个密集播放视频片段。
  23. 根据权利要求22所述的服务器，其中，所述处理器从视频中选择至少一个中选视频帧还包括：确定在每个密集播放视频片段中抽取的中选视频帧的个数。
  24. 根据权利要求23所述的服务器，其中，所述处理器确定在每个密集播放视频片段中抽取的中选视频帧的个数，包括：计算每个密集播放视频片段中所有视频帧的平均播放密集程度；计算所述每个密集播放视频片段中播放密集程度大于或等于所述平均播放密集程度的视频帧所占比例；以及根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
  25. 根据权利要求24所述的服务器，其中，所述处理器根据所述比例，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数，包括：根据所述比例从大到小，逐步减少在所述每个密集播放视频片段中抽取的中选视频帧的个数。
  26. 根据权利要求23所述的服务器，其中，所述处理器确定在每个密集播放视频片段中抽取的中选视频帧的个数，包括：确定每个密集播放视频片段的长度；以及根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数。
  27. 根据权利要求26所述的服务器，其中，所述处理器根据所述每个密集播放视频片段的长度，确定在所述每个密集播放视频片段中抽取的中选视频帧的个数包括：根据所述每个密集播放视频片段的长度从长到短，逐步减少在所述每 个密集播放视频片段中抽取的中选视频帧的个数。
  28. 一种计算机可读存储介质，其特征在于，包括计算机执行指令，以供计算机的处理器执行所述计算机执行指令时，所述计算机执行如权利要求1至13中任一项所述的产生视频帧集合的方法。
PCT/CN2015/086493 2014-10-31 2015-08-10 产生视频帧集合的方法、设备及服务器 WO2016065972A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/522,546 US10313712B2 (en) 2014-10-31 2015-08-10 Method, device, and server for producing video frame set

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410610673.9A CN105635749B (zh) 2014-10-31 2014-10-31 产生视频帧集合的方法和设备
CN201410610673.9 2014-10-31

Publications (1)

Publication Number Publication Date
WO2016065972A1 true WO2016065972A1 (zh) 2016-05-06

Family

ID=55856568

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/086493 WO2016065972A1 (zh) 2014-10-31 2015-08-10 产生视频帧集合的方法、设备及服务器

Country Status (3)

Country Link
US (1) US10313712B2 (zh)
CN (1) CN105635749B (zh)
WO (1) WO2016065972A1 (zh)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107094269A (zh) * 2016-11-04 2017-08-25 口碑控股有限公司 数据处理方法、设备和视频播放装置
CN107197378A (zh) * 2017-06-23 2017-09-22 深圳天珑无线科技有限公司 一种视频信息的处理方法及装置
CN109756767B (zh) * 2017-11-06 2021-12-14 腾讯科技（深圳）有限公司 预览数据播放方法、装置及存储介质
CN110121115B (zh) * 2018-02-06 2023-02-10 阿里巴巴（中国）有限公司 精彩视频片段的确定方法及装置
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
CN111277861B (zh) * 2020-02-21 2023-02-24 北京百度网讯科技有限公司 提取视频中热点片段的方法以及装置
US11350162B2 (en) 2020-05-05 2022-05-31 Rovi Guides, Inc. Systems and methods to determine reduction of interest in a content series
US11153643B1 (en) * 2020-05-05 2021-10-19 Rovi Guides, Inc. Systems and methods to determine reduction of interest in a content series

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101783915A (zh) * 2010-03-19 2010-07-21 北京国双科技有限公司 一种实现视频量化的方法
CN103190156A (zh) * 2010-09-24 2013-07-03 株式会社Gnzo 视频比特流的传输系统
CN103957433A (zh) * 2014-03-31 2014-07-30 深圳市同洲电子股份有限公司 一种视频数据的处理方法、相关设备及系统
WO2014132987A1 (ja) * 2013-02-27 2014-09-04 ブラザー工業株式会社 情報処理装置及び情報処理方法

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5623344A (en) * 1992-09-01 1997-04-22 Hitachi America, Ltd. Digital video recording device with trick play capability
US6377748B1 (en) * 1997-02-18 2002-04-23 Thomson Licensing S.A. Replay bit stream searching
KR101146926B1 (ko) * 2006-12-20 2012-05-22 엘지전자 주식회사 이동 단말기에서 비디오의 대표 영상 제공 방법
CN101901619B (zh) * 2010-07-16 2012-10-17 复旦大学 一种基于视频内容缩影的增强用户体验的视频播放器
CN102447973B (zh) * 2011-10-10 2013-12-04 华为技术有限公司 一种缓存调整的方法、装置和系统
US9542976B2 (en) * 2013-09-13 2017-01-10 Google Inc. Synchronizing videos with frame-based metadata using video content
CN103634605B (zh) 2013-12-04 2017-02-15 百度在线网络技术（北京）有限公司 视频画面的处理方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101783915A (zh) * 2010-03-19 2010-07-21 北京国双科技有限公司 一种实现视频量化的方法
CN103190156A (zh) * 2010-09-24 2013-07-03 株式会社Gnzo 视频比特流的传输系统
WO2014132987A1 (ja) * 2013-02-27 2014-09-04 ブラザー工業株式会社 情報処理装置及び情報処理方法
CN103957433A (zh) * 2014-03-31 2014-07-30 深圳市同洲电子股份有限公司 一种视频数据的处理方法、相关设备及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WU, QIAN ET AL.: "Brief Analysis of Video Abstract Technology", JOURNAL OF COMMUNICATION UNIVERSITY OF CHINA ( SCIENCE AND TECHNOLOGY, vol. 15, no. 2, 30 June 2008 (2008-06-30), pages 56 *

Also Published As

Publication number Publication date
US20170318317A1 (en) 2017-11-02
CN105635749A (zh) 2016-06-01
US10313712B2 (en) 2019-06-04
CN105635749B (zh) 2017-03-22

Similar Documents

Publication Publication Date Title
WO2016065972A1 (zh) 产生视频帧集合的方法、设备及服务器
US9736503B1 (en) Optimizing timing of display of a mid-roll video advertisement based on viewer retention data
US9736432B2 (en) Identifying popular network video segments
US9807466B2 (en) Managing interactive subtitle data
JP5086189B2 (ja) 動画コンテンツのダイジェスト映像を生成するサーバ、方法及びプログラム
CN106454431B (zh) 电视节目推荐方法和系统
WO2019144838A1 (zh) 一种用于获得视频的评价结果信息的方法和装置
WO2014022200A1 (en) Ad selection and next video recommendation in a video streaming system exclusive of user identity-based parameter
US20170289226A1 (en) Video analytics device
CN111512635A (zh) 用于选择性跳过媒体内容的方法和系统
CN111385606A (zh) 一种视频预览方法、装置及智能终端
CN106791930B (zh) 一种视频处理方法和装置
CN109089169A (zh) 一种直播间切换方法、装置及存储介质
CN110418191A (zh) 一种短视频的生成方法及装置
US20120063746A1 (en) Method and apparatus for extracting key frames from a video
CN104768073A (zh) 一种频道菜单的显示方法及装置
CN112653918A (zh) 预览视频生成方法、装置、电子设备及存储介质
CN113630630A (zh) 一种视频解说配音信息的处理方法、装置及设备
CN111277898A (zh) 一种内容推送方法及装置
JP2006054747A (ja) 情報処理装置および方法、並びにプログラム
CN112492351A (zh) 一种视频处理方法、装置、设备及存储介质
CN113014981A (zh) 视频播放方法、装置、电子设备及可读存储介质
CN104410874A (zh) 视频粘度信息的检测方法、装置和系统
CN112055233B (zh) 基于广告收视率控制广告播放方法及装置
CN107197378A (zh) 一种视频信息的处理方法及装置

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: 15854101

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 15522546

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15854101

Country of ref document: EP

Kind code of ref document: A1