CN103501434A - Method and device for analyzing quality of video - Google Patents
Method and device for analyzing quality of video Download PDFInfo
- Publication number
- CN103501434A CN103501434A CN201310425590.8A CN201310425590A CN103501434A CN 103501434 A CN103501434 A CN 103501434A CN 201310425590 A CN201310425590 A CN 201310425590A CN 103501434 A CN103501434 A CN 103501434A
- Authority
- CN
- China
- Prior art keywords
- video
- operation behavior
- user
- information
- time
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Images
Abstract
The invention provides a method and device for analyzing the quality of a video, and aims to solve the problems that no uniform standard is available for evaluating the quality of video resources, and evaluation of the quality is not objective. The method includes the steps that in the video playing process, operation behavior information of each user in unit time of the video is detected; regarding each unit time, the operation behavior information of the users is weighed, and then the skipping rate of the video in the unit time is determined; quality evaluation information of the video is determined according to the skipping rate of the video in the unit time. According to the method and device for analyzing the quality of the video, the quality of the video is determined by analyzing a large number of operation behaviors of the users, and evaluation of the quality is accurate and objective.
Description
Technical field
The present invention relates to network technology, particularly relate to a kind of mass analysis method and device of video.
Background technology
The streaming media resources such as current network middle pitch, video are numerous, particularly video resource, not only have the movie and television play resources such as film, TV, variety show in network, also has micro-film, the resource such as shoot the video certainly, but in network, the quality of video content is uneven, and there is no unified standard.
Take the movie and television play resource as example, usually the making of movie and television play and distribution team can be publicized the movie and television play content by media, video website etc. are understood the quality of movie and television play content by media, then determine whether to buy the copyright of this movie and television play, the user is also by the publicity of media and the recommendation of video website, judge the quality of movie and television play, to determine whether to watch this movie and television play.
Therefore, the copyright of current movie and television play is sold and audience ratings, basically depends on the propaganda strength of this movie and television play.Thereby it is fine that the copyright that a lot of movie and television plays occurred is sold situation, but movie and television play poor quality in fact causes calmly the situation that audience ratings is very low, causes the waste of waste video website resource, and user bandwidth resource and waste of time.Meanwhile, although also have some movie and television play quality better, owing to making and publicizing, less input, and by the user, do not understood, thereby cause the idle waste problem of resource.
Why there will be the problems referred to above, main cause is that the quality evaluation to video resource does not have unified standard, and the evaluation of quality is not objective.
Summary of the invention
The invention provides a kind of mass analysis method and device of video, with the quality evaluation solved video resource, there is no unified standard, and the evaluation of quality one sided problem.
In order to address the above problem, the invention discloses a kind of mass analysis method of video, comprising:
In video display process, detect respectively constituent parts operation behavior information time in of each user at video;
For each unit interval, by the weighted calculation of the operation behavior information to each user, determine the skip rate of video in the described unit interval;
Determine the quality evaluation information of described video in the skip rate of constituent parts time according to described video.
Preferably, in described video display process, detect respectively constituent parts operation behavior information time in of each user at video, comprising: in video display process, detect each user's operation behavior and the temporal information of described operation behavior; Described video was divided according to the unit interval, and the operation behavior by constituent parts in the time and the temporal information of described operation behavior form the operation behavior information of user within the described unit interval.
Preferably, in described video display process, detect each user's operation behavior and the temporal information of described operation behavior, comprise: in video display process, detect each user's all types of operation behavior, wherein, the type of described operation behavior comprises following at least one: F.F., rewind down, drag, stop to play and close player; Record the temporal information of the time of origin of operations behavior as described operation behavior; And when the type of described operation behavior is F.F., rewind down or while dragging, the concluding time of recording described operation behavior, and the described concluding time is added in the temporal information of described operation behavior.
Preferably, described for each unit interval, by the weighted calculation of the operation behavior information to each user, determine the skip rate of video in the described unit interval, comprise: within each unit interval of video, the operation behavior information to each user within the described unit interval is added up; According to preset weight parameter, operation behavior information is weighted to calculating, determines the skip rate of video in the described unit interval.
Preferably, determine the quality evaluation information of described video in the skip rate of constituent parts time according to described video, comprise: described video is compared in the skip rate of constituent parts time and the preset thresholding of skipping, and determine that according to comparative result the play quality value is as quality evaluation information.
Preferably, also comprise: described video is divided according to programme content, and determined the reproduction time of each section programme content; According to described video, determine the quality evaluation information of described video in the skip rate of constituent parts time, comprise: according to the time, the skip rate of constituent parts time and described programme content are carried out associatedly, determine that the resource content evaluation information of described video is as quality evaluation information.
Preferably, also comprise: obtain user profile, and classified according to described user's information; The programme content of the lower user of each classification of the statistics skip rate in the time and association thereof at constituent parts respectively, determine and user's evaluation information of described video to described user's evaluation information added in described quality evaluation information.
Accordingly, the present invention also provides a kind of quality evaluation device of video, comprising:
The behavior detection module, at video display process, detect respectively constituent parts operation behavior information time in of each user at video;
The skip rate determination module, for for each unit interval, by the weighted calculation of the operation behavior information to each user, determine the skip rate of video in the described unit interval;
The quality determination module, for determining the quality evaluation information of described video according to described video in the skip rate of constituent parts time.
Preferably, described behavior detection module comprises: detection sub-module, at video display process, detect each user's operation behavior and the temporal information of described operation behavior; Divide and form submodule, for described video was divided according to the unit interval, the operation behavior by constituent parts in the time and the temporal information of described operation behavior form the operation behavior information of user within the described unit interval.
Preferably, described detection sub-module comprises: the operation behavior detecting unit, be used at video display process, detect each user's all types of operation behavior, wherein, the type of described operation behavior comprises following at least one: F.F., rewind down, drag, stop to play and close player; Record cell, the temporal information for the time of origin that records the operations behavior as described operation behavior; And when the type of described operation behavior is F.F., rewind down or while dragging, the concluding time of recording described operation behavior, and the described concluding time is added in the temporal information of described operation behavior.
Preferably, described skip rate determination module comprises: the statistics submodule, and within each unit interval of video, the operation behavior information to each user within the described unit interval is added up; Calculating sub module, be weighted calculating for the weight parameter according to preset to operation behavior information, determines the skip rate of video in the described unit interval.
Preferably, described quality determination module comprises: mass value is estimated submodule, for described video is compared in the skip rate of constituent parts time and the preset thresholding of skipping, and determines that according to comparative result the play quality value is as quality evaluation information.
Preferably, also comprise: the division of teaching contents module, for described video is divided according to programme content, and the reproduction time of definite each section programme content; The quality determination module comprises: the resource content evaluation submodule, and associated for according to the time, the skip rate of constituent parts time and described programme content being carried out, determine that the resource content evaluation information of described video is as quality evaluation information.
Preferably, also comprise: user's sort module, for obtaining user profile, and classified according to described user's information; The quality determination module also comprises: the user estimates submodule, the programme content of lower the user skip rate in the time and association thereof at constituent parts for each classification of statistics respectively, determine user's evaluation information of described video, described user's evaluation information is added in described quality evaluation information.
Compared with prior art, the present invention includes following advantage:
Detect user's operation behavior information in video display process, take each user to the real behavior of video-see as basis, can embody truly user's evaluation.Then for the weighted calculation of each unit interval by the operation behavior information to each user, determine the skip rate of video in the described unit interval, and then the quality evaluation information of definite described video, determine the quality of video by the operation behavior of analyzing a large number of users, quality evaluation is more accurate, objective.
The accompanying drawing explanation
Fig. 1 is the mass analysis method flow chart of the video that provides of the embodiment of the present invention one;
Fig. 2 is the mass analysis method flow chart of the video that provides of the embodiment of the present invention two;
Fig. 3 is that in embodiment, the Hooke based on skip rate of first video is listed as civilian curve;
Fig. 4 is that in embodiment, the Hooke based on skip rate of second video is listed as civilian curve;
Fig. 5 is that the Hooke based on skip rate of the 3rd video in embodiment is listed as civilian curve;
Fig. 6 is the quality analysis apparatus structure chart of the video that provides of the embodiment of the present invention three.
Embodiment
For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, below in conjunction with the drawings and specific embodiments, the present invention is further detailed explanation.
The embodiment of the present invention provides a kind of mass analysis method of video, detects user's operation behavior information in video display process, take each user to the real behavior of video-see as basis, can embody truly user's evaluation.Then for the weighted calculation of each unit interval by the operation behavior information to each user, determine the skip rate of video in the described unit interval, and then the quality evaluation information of definite described video, determine the quality of video by the operation behavior of analyzing a large number of users, quality evaluation is more accurate, objective.
Embodiment mono-
With reference to Fig. 1, provided the mass analysis method flow chart of the video that the embodiment of the present invention one provides.
The embodiment of the present invention is for the quality of accurate evaluation video, and the operation behavior while based on the user, watching video is analyzed.User's operation behavior can embody the interest level of user to video, if for example the user loses interest in to certain section content, may select F.F. or skip, if interesting to certain section content, may retreat again and watch.The user is the evaluation to video on the whether video interested audience ratings that can directly affect video and user, thereby can analyze the quality of video by user's operation behavior.
Therefore watching video the user, is also in the playing process of video, detects the operation behavior of user to video.Video can be divided into to some unit interval in actual treatment, then detect the operation that in each unit interval, the user carries out and form operation behavior information.Wherein, the unit interval can be set according to actual demand, as 1 second, and 1 minute or 1 frame etc.User's operation behavior refers to user's play operation to video when watching video, as F.F., drag, stop to play etc.
Then for each unit interval of this video, collect and allly watch the user of this video in the operation behavior information of this unit interval, as played 100 person-times of video, 90 person-times of F.F.s are wherein arranged in the time of the 10th second, in the time of the 50th second, there are 20 person-times to retreat etc.Thereby by the operation behavior to a large number of users, analyzed, the real user that can obtain having Statistical Value watches evaluation.
The embodiment of the present invention is each generic operation behavior configuration respective weights, thereby all users' operation behavior information is weighted according to weight separately, by calculating as operation behavior on average being arrived to each user etc., determine the skip rate of video in this unit interval.Described skip rate refers to the probability that video is skipped when playing, and skips and refers to the content that the user did not watch or watched fast this unit interval.
Can obtain the user by this skip rate and select in this unit interval of video the probability of not watching or watching fast, and then the embodiment user is to the interest level of this unit interval play content, if skip rate is higher, illustrate that the user is not interested to this unit interval play content, this section content quality is lower; If skip rate is lower, illustrate that this unit interval play content of user is more interesting, usually can watch and even repeatedly watch this section, quality is higher.Thereby can obtain the interest-degree of user to video content under each period by video in the skip rate of constituent parts time, and then the quality of analyzing video obtains quality evaluation information.This quality evaluation information can be based on the mass value that skip rate obtains, and as skip rate, below 10%, video quality value is that the 9(full marks are 10); Can be also that skip rate is associated with type or the actual content of this unit interval of video, thus the quality of the type based on video or resource content evaluation video.
In sum, detect user's operation behavior information in video display process, take each user to the real behavior of video-see as basis, can embody truly user's evaluation.Then for the weighted calculation of each unit interval by the operation behavior information to each user, determine the skip rate of video in the described unit interval, and then the quality evaluation information of definite described video, determine the quality of video by the operation behavior of analyzing a large number of users, quality evaluation is more accurate, objective.
Embodiment bis-
With reference to Fig. 2, provided the mass analysis method flow chart of the video that the embodiment of the present invention two provides.
On the basis of above-described embodiment, the present embodiment is discussed the mass analysis method of video in detail, specifically comprises the steps:
Step 201, in video display process, detect each user's operation behavior and the temporal information of described operation behavior.
In optional embodiment of the present invention, in described video display process, detect each user's operation behavior and the temporal information of described operation behavior, comprising: in video display process, detect each user's all types of operation behavior; Record the temporal information of the time of origin of operations behavior as described operation behavior; And when the type of described operation behavior is F.F., rewind down or while dragging, the concluding time of recording described operation behavior, and the described concluding time is added in the temporal information of described operation behavior.
Operation behavior to each user in the playing process of video is detected, such as can in the player of video, configuring the detectors such as the process that detected or thread, to detect the play operation of user to video, use as shortcut, dragging etc. of progress bar, this player is that the device of displaying video is as the player client in computer or be embedded in player in webpage etc.Wherein, the type of described operation behavior comprises following at least one: F.F., rewind down, drag, stop to play and close player.
Wherein, drag broadcasting dragging and dragging backward forward, can be by the rising of operation, only the time is determined, dragging with F.F. forward is similar, is to skip or play fast certain section content; Dragging with rewind down backward is similar, refers to and returns repeat playing section content.Therefore also need F.F. and drag, and rewind down is distinguished with dragging, operation that can User in actual treatment is distinguished, as F.F. or fast reverse operation are that fast forward button (or shortcut) from player is obtained, and drag operation is to obtain from the operation to the player plays bar; Or distinguished according to the length of operating time, as within 10 seconds, play with interior (or repetition) fast, be F.F. (or rewind down), surpass to be in 10 seconds and drag, concrete distinguishing condition can be set according to actual demand, and the embodiment of the present invention is not construed as limiting this.
Stop playing and refer to video playback and stop, can not comprising the operation of time-out; Close player and refer to the broadcast source of closing video, close the operation of this player while being play as employing player client, close the operation of webpage while and for example adopting webpage to play.
Record the time of origin of this operation behavior and this operation behavior when user's operation behavior being detected, as beginning F.F. in the 5th minute 38 seconds, and for example within the 56th minute 08 second, stop playing, the time of origin of operation behavior can be recorded as to the temporal information of operation behavior.Wherein, due to F.F., rewind down or to drag be the operation behavior of continuation, be that the type is to maintain a period of time, therefore to record this operation behavior concluding time to this generic operation simultaneously, as forwarding operation within the 5th minute 38 seconds to 5 minutes and 43 seconds, detected, then the concluding time of this generic operation behavior is also joined in the temporal information of this operation behavior, thereby can count duration of the operation behavior of this class continuation.
And, when detecting F.F., rewind down and drag operation, the minimum time thresholding that starts statistics also to be set, i.e. F.F., rewind down or drag and continue how many seconds and record the once-through operation behavior, as 1 second or at least one unit interval etc.
Step 202, divided described video according to the unit interval, the operation behavior by constituent parts in the time and the temporal information of described operation behavior form the operation behavior information of user within the described unit interval.
Then video was divided according to the unit interval, as take 1 second as the unit interval, one section video of 30 minutes just comprises 1800 unit interval, then the above-mentioned operation behavior detected is divided according to temporal information, counts on respectively in the operation behavior information of constituent parts under the time.As a certain user detected and carried out forwarding operation at the 5th minute 38 seconds to 5 minutes and 43 seconds, in its operation behavior information within 338 seconds to 343 seconds these 5 unit interval, comprise respectively the temporal information of forwarding operation and this operation.
Step 203, within each unit interval of video, the operation behavior information to each user within the described unit interval is added up.
After to each user, at constituent parts, the operation behavior information in the time is all added up, is divided, take the unit interval as benchmark, all users that watch this video of each of video statistics when played (be video at every turn), the number of times of all kinds of operation behavior information detected unit interval.According to the type of operation behavior, added up, thereby can count under the broadcasting total degree of video, the number of times that in each unit interval, all kinds of operation behaviors are carried out, be 100 times as play total degree, the number of times of this unit interval F.F. in the 338th second 68 times, the number of times that stops playing is 6 times.
Step 204, be weighted calculating according to preset weight parameter to operation behavior information, determines the skip rate of video in the described unit interval.
The embodiment of the present invention is each generic operation behavior configuration respective weights, can be within each unit interval, the number of times of all types of operation behaviors is weighted according to weight separately, then add up the weighted sum of all operations behavior, calculate the skip rate of this unit interval, in weighted sum is on average play to video at every turn, be about to the broadcasting total degree of weighted sum divided by video, the probability of skipping while obtaining at every turn playing.
And, F.F. and rewind down, dragging backward with dragging forward is operation behavior respect to one another, therefore when determining weight, can will be configured to identical weight, add F.F. while calculating weighted sum or the weighted value of the number of times that drags backward, deduct rewind down or the authentication values of the number of times that drags forward; Perhaps be configured to opposite number, wherein F.F. or the weight that drags backward are positive number, and rewind down or the weight dragged forward are negative, calculating weighted sum, are that weighted value addition is separately got final product.Above-mentioned is only a kind of weight configuration embodiment, also can the different weight of each self-configuring, and the embodiment of the present invention is not construed as limiting this.
Because F.F., rewind down and drag operation are the continuation operations, therefore the length of this operation duration also can produce certain impact to weight, it is the duration difference of same operation behavior, its weight may be different, for example, take 15 seconds as critical point, F.F. less than 15 seconds (containing 15 seconds) is remembered a weights N1, and user's F.F. is within 15 seconds, remembering another weights N2.Therefore also can be according to the difference difference statistics number of duration.
For example, adopt Hooke row literary composition value to calculate the skip rate of a certain unit interval, wherein by F.F. or drag backward, with rewind down or drag forward and be configured to identical weight according to the duration and give an example, following parameter be set:
(1) operation behavior is F.F. or while dragging backward, will start F.F./drag from this unit interval, and the operation behavior that and F.F./drag duration is less than or equal to 15 seconds is designated as A1; Start F.F./drag from this unit interval, the operation behavior that and F.F./drag duration is greater than 15 seconds is designated as A2; Start F.F./drag from the other unit time, this unit interval (be F.F./drag through or arrive this unit interval) in the scope of F.F./drag, the duration of and F.F./drag is less than or equal to the operation behavior of 15 seconds and is designated as A3; Start F.F./drag from the other unit time, the duration of and F.F./drag was greater than the operation behavior of 15 seconds and was designated as A4 in the scope of F.F./drag this unit interval.
(2) operation behavior is play for stopping, and shut-down operation can comprise and starts to play other videos or enter the behaviors such as web site contents at this page, and this operation behavior is recorded as to B.
(3) operation behavior when closing player, comprises and stops accessing this website,, this operation behavior is recorded as to C.
(3) operation behavior is rewind down or while dragging forward, will start rewind down/drag from this unit interval, and the operation behavior that and rewind down/drag duration is less than or equal to 15 seconds is designated as D1; Start rewind down/drag from this unit interval, the operation behavior that and rewind down/drag duration is greater than 15 seconds is designated as D2; Start rewind down/drag from the other unit time, this unit interval (be rewind down/drag through or arrive this unit interval) in the scope of rewind down/drag, the duration of and rewind down/drag is less than or equal to the operation behavior of 15 seconds and is designated as D3; Start rewind down/drag from the other unit time, the duration of and rewind down/drag was greater than the operation behavior of 15 seconds and was designated as D4 in the scope of rewind down/drag this unit interval.
(5) the video playback total degree is recorded as to M.
Skip rate=
(number of times of number of times-N4*D4 of number of times-N3*D3 of number of times-N2*D2 of number of times-N1*D1 of number of times+N6*C of number of times+N5*B of number of times+N4*A4 of number of times+N3*A3 of number of times+N2*A2 of N1*A1)/M
Wherein, in video display process, certain unit interval user may carry out rewind down, forwarding operation repeatedly, therefore above-mentioned F.F./drag, and the number of times of and rewind down/drag statistics may be greater than the total degree of video playback.And the operation behavior of above-mentioned only several types, may also comprise the operation behavior of other types in actual treatment, do not enumerate herein, should not be understood as the restriction to the embodiment of the present invention, for the operation behavior of not enumerating, calculate and get final product according to the method described above.
Step 205, determine the quality evaluation information of described video according to described video in the skip rate of constituent parts time.
While according to skip rate, determining the quality evaluation information of video, can adopt following at least one mode:
(1) described video is compared in the skip rate of constituent parts time and the preset thresholding of skipping, and determine that according to comparative result the play quality value is as quality evaluation information.
In the embodiment of the present invention, can pre-configuredly skip thresholding, this skip thresholding can for the user of calculating and setting average skip the probability threshold value, thereby the skip rate of constituent parts time and the preset thresholding of skipping are compared, and for example, the skip rate of statistical unit time is skipped the duration below thresholding at this, determine that this duration accounts for the ratio of video total time, the mass value of the corresponding different brackets of different ratios, mass value as higher as ratio is higher, using mass value as quality evaluation information.And for example, for skip rate configures corresponding mass value, calculate the average skip rate of this video, skip rate with value divided by total number of unit interval, thereby determine mass value corresponding to average skip rate, as quality evaluation information.
(2) described video is divided according to programme content, and determined the reproduction time of each section programme content; Associated according to the time by the skip rate of constituent parts time and described programme content, determine that the resource content evaluation information of described video is as quality evaluation information.
Programme content refers to the content of playing in video, also video can be divided according to programme content, determine the reproduction time that each programme content is corresponding, then according to the time, that programme content is associated with the skip rate of corresponding unit interval, as from starting by 10 minutes as programme content 1, if time in seconds, that programme content 1 is associated with the skip rate of front 600 unit interval, thereby can obtain the skip rate in this programme content period, and then the resource content evaluation information based on programme content of obtaining is as quality evaluation information.As this section programme content, for making laughs or the comedy type, skip rate is lower, and resource content evaluation information can be preferred for the user program of comedy type, or the program audience rating of comedy type is more high.
(3) obtain user profile, and classified according to described user's information; The programme content of the lower user of each classification of the statistics skip rate in the time and association thereof at constituent parts respectively, determine and user's evaluation information of described video to described user's evaluation information added in described quality evaluation information.
Can also obtain user profile when analyzing video quality, this user profile can be that user's log-on message is as sex, age, occupation etc., the history that can be also the user is play record etc., then User information is classified to the user, as according to character classification by age, and for example according to the type of watching video, classified etc.Then the lower user of each classification of statistics at constituent parts the skip rate in the time, can also further add up and the corresponding associated programme content of this skip rate, thereby determine the user's evaluation information based on the user, described user's evaluation information is added in described quality evaluation information.
For example, determine that the each age group user watches the mass value of a certain video, thereby obtain this video, relatively be applicable to liking of which age bracket user, in this section age user, quality evaluation is higher.And for example, determine and often see that the user of terrible class video is interested in the content X of video 1, often see that the user of scientific and technological class video is interested in the content Y of video 2, thereby obtain the different quality evaluation information of video.
Therefore, in the embodiment of the present invention, quality evaluation information can be the numerical value of a quantification, can also further estimate type, the programme content of video, audient all (user) etc., thus obtain comprehensive quality evaluation.
Have the methods of marking of some quantized values in prior art, but normally the user gives a mark voluntarily, the evaluation criterion disunity, and have certain customers may watch but not marking of video, so data are not comprehensive.And the embodiment of the present invention is according to the operation behavior of user in video playback, the user has watched this video to be added up, and data are more comprehensive.And statistics, analysis by a large amount of operation behaviors, have unified standards of grading, the mass value of the quantification therefore obtained is more accurate, also more objective.
The above-mentioned skip rate according to the constituent parts time has been determined the quality evaluation information of video, can also be by mode content and quality evaluations of display video more clearly such as drawing in actual treatment.To as above civilian value of example employing Hooke row calculating skip rate, the corresponding Hooke that can also draw this skip rate is listed as civilian curve, as shown in Fig. 3,4 and 5:
Fig. 3 provides the Hooke based on skip rate of first video to be listed as civilian curve, and wherein transverse axis is the time, and the longitudinal axis is skip rate, and the programme content that each label is corresponding is: 3.01,1:01 prologue dance; 3.02, what Gui of 1:37 says; 3.03,3:33 thanks to Na and learns Liu Huan; 3.04, the 4:28 dance of beginning; 3.05, introduce the welcome guest; 3.06, the 11:11 Chen Long VCR that rehearses; 3.07,11:38 Chen Long performance; 3.08,13:35 Chen Long performance; 3.09, the 14:10 groups of stars imitate; 3.10, welcome guest comment; 3.11, the fragrant VCR that makes laughs of the yellow skill of 18:10; 3.12, the fragrant performance of the yellow skill of 19:30; 3.13, Huang Yixin imitates the cat king; 3.14, thank to Na and make laughs fragrant what Gui of the yellow skill of 22:39; 3.15, the 25:39 first round gives a mark; 3.16, Shen Ling Jia Ling freely circles in the air; 3.17, your VRS that rehearses of 33:04 Li Xin; 3.18, your different energy of 33:34 Li Xin; 3.19,34:47 Li Xin you the performance; 3.20,35:35 Li Xin you dance; 3.21,38:10 thanks to Na and imitates Wu great Lang; 3.22, the performance of 3.2241 minutes Shao Mei fine jades; 3.23,42:45 performance recumbency plays the piano; 3.24,45:51 second takes turns marking; 3.25, Xie Na performs Wu great Lang; 3.26, the Qu Ying roc imitates the phoenix legend; 3.27, Shen Ling imitates that once be lost can; 3.28,58:20 opens auspicious imitation VCR; 3.29,58:20 opens auspicious performance; 3.30,63:10 king ancestral Bluepoint comments; 3.31,65:22 Anlin holds " discriminating Huan passes " fragment; 3.32,66 Anlin hold rehearsal VCR; 3.33, the imitative Li Yuchun of 67:24 Anlin molar; 3.34,68:49 Anlin hold to dance; 3.35,71:24 third round marking; 3.36, the bright singing of 73:10 Malaysian; 3.37,76:01 thanks Na, the bright dancing of what Gui comment Malaysian; 3.38, in what Gui nonsense of 76:33; 3.39,81:50 finally gives a mark; 3.40,84:55 makes up and makes laughs outside the venue.
This video is " changeable large coffee show " the 2013-01-10 phase, take this video as example, if using 40% as the threshold value of skipping probability, in the skip rate of constituent parts time, basically all in this below threshold value, mass value can be 9 minutes (full marks are 10 minutes) to this video.Specific to programme content and associated unit interval, skip probable value low especially (for bearing) during as Jia Ling and Shen Ling chorus " freely circling in the air ", obtain quality evaluation information as: 1, Jia Ling and Shen Ling are more welcome; 2, the class of making laughs is imitated more welcome.Thereby, by comprehensive statistics, the analysis to above-mentioned data, can draw the following quality evaluation information of this video: 1, " changeable large coffee show " whole program quality is pretty good; 2, such program class that is applicable to making laughs is imitated; 3, the class of making laughs is imitated the welcome guest Jia Ling and Shen Ling is arranged preferably.
This quality evaluation information can also feed back to the making side of video etc., can choose subject matter, scene and welcome guest with reference to this quality evaluation information when follow-up making, and purchaser or content play side also can be according to quality evaluation Information Selection screens, and recommend video etc. for the user.
Fig. 4 provides the Hooke based on skip rate of second video to be listed as civilian curve, and wherein transverse axis is the time, and the longitudinal axis is skip rate, and the programme content that each label is corresponding is: 4.01, classical old song skewered; 4.02, host's Chu Chang & Interview Yang Yu jade-like stone; 4.03, Yang Yuying " tea hill love song " newly sings; 4.04, sing; 4.05, Fang Tan & Hair is peaceful to appear on the scene; 4.06, host's interaction; 4.07, hair peaceful " never forgetting ", " legend "; 4.08, Chen Shaohua " 99 Kinawa "; 4.08, Huang Gexuan " like in late autumn " " silence is golden "; 4.10, Wang Han chorus; 4.11, Chen Ming " Shanghai " " Our Time to Shine " " time passes swiftly like flowing water " " waiting you to like me "; 4.12, conclusion.
This video is first phase " making progress every day " program, and wherein, while dividing according to programme content, can there be overlapping phenomenon in the part unit interval.From curve, passable all this videos are when being played to old song and singing, and user's skip rate is all higher, and classical old song is not liked by the main body audient colony that can to obtain quality evaluation information be " making progress every day " program.And then User classification is segmented the user group of this program, as the analysis by the user group, the audient colony that obtains " making progress every day " mostly is women after 90s, thereby can draw separately and take the Leeuwenhoek curve that women after 90s is target audience colony, allow curve only react the viewing behavior statistics of target audience colony.
Fig. 5 provides the Hooke based on skip rate of the 3rd video to be listed as civilian curve, and wherein transverse axis is the time, and the longitudinal axis is skip rate, and the programme content that each label is corresponding is: 5.01, host's opening remarks; 5.02, advertisement; 5.03, the boy student sings; 5.04, boy student's talk of leaving the theatre; 5.05, the good yellow Han of Meng Feile tries sunglasses on; 5.06, male welcome guest dubs to Journey to the West; 5.07, the boy student finally chooses; 5.08, boy student's right.
The skip rate that can obtain the time spot from above-mentioned curve is all higher, and can delete this partial data according to programme content in actual treatment, and the embodiment of the present invention is not construed as limiting this.And boy student's skip rate of talking of leaving the theatre is also very high.It analyze can be formed to quality evaluation information feed back to program making side and content play side provides detailed user watched behavior reference.
In sum, by all types of operation behaviors and temporal information thereof to the user, added up, can obtain the operation behavior information in the constituent parts time, the type of operation behavior comprises following at least one: F.F., rewind down, drag, stop to play and close player, the type of operation behavior is various, can improve the accuracy of data.And operation behavior can be definite according to the weight of dissimilar correspondence weighted value, and then the skip rate of unit of account time, make the more accurate of skip rate.
Secondly, the quality evaluation information of video can be based on the mass value of skip rate, is a quantized value that standard is unified.Also can be further associated by the skip rate of unit interval and programme content, obtain content-based quality evaluation information, can also determine skip rate and associated programme content thereof that the user classified in the lower unit interval, obtain the quality evaluation information based on the user, thereby make quality evaluation information more comprehensive.
Embodiment tri-
Based on above-described embodiment, the present embodiment also provides a kind of quality analysis apparatus of video.
With reference to Fig. 6, provided the quality analysis apparatus structure chart of the video that the embodiment of the present invention three provides.
The quality analysis apparatus of this video comprises: behavior detection module 601, skip rate determination module 602 and quality determination module 603.
Wherein: behavior detection module 601, at video display process, detect respectively constituent parts operation behavior information time in of each user at video;
Skip rate determination module 602, for for each unit interval, by the weighted calculation of the operation behavior information to each user, determine the skip rate of video in the described unit interval;
Quality determination module 603, for determining the quality evaluation information of described video according to described video in the skip rate of constituent parts time.
In sum, detect user's operation behavior information in video display process, take each user to the real behavior of video-see as basis, can embody truly user's evaluation.Then for the weighted calculation of each unit interval by the operation behavior information to each user, determine the skip rate of video in the described unit interval, and then the quality evaluation information of definite described video, determine the quality of video by the operation behavior of analyzing a large number of users, quality evaluation is more accurate, objective.
In optional embodiment of the present invention, described behavior detection module comprises: detection sub-module, at video display process, detect each user's operation behavior and the temporal information of described operation behavior; Divide and form submodule, for described video was divided according to the unit interval, the operation behavior by constituent parts in the time and the temporal information of described operation behavior form the operation behavior information of user within the described unit interval.
In optional embodiment of the present invention, described detection sub-module, comprise: the operation behavior detecting unit, be used at video display process, detect each user's all types of operation behavior, wherein, the type of described operation behavior comprises following at least one: F.F., rewind down, drag, stop to play and close player; Record cell, the temporal information for the time of origin that records the operations behavior as described operation behavior; And when the type of described operation behavior is F.F., rewind down or while dragging, the concluding time of recording described operation behavior, and the described concluding time is added in the temporal information of described operation behavior.
In optional embodiment of the present invention, described skip rate determination module comprises: the statistics submodule, and within each unit interval of video, the operation behavior information to each user within the described unit interval is added up; Calculating sub module, be weighted calculating for the weight parameter according to preset to operation behavior information, determines the skip rate of video in the described unit interval.
In optional embodiment of the present invention, described quality determination module, comprise: mass value is estimated submodule, for described video is compared in the skip rate of constituent parts time and the preset thresholding of skipping, and determines that according to comparative result the play quality value is as quality evaluation information.
In optional embodiment of the present invention, also comprise: the division of teaching contents module, for described video is divided according to programme content, and the reproduction time of definite each section programme content; The quality determination module comprises: the resource content evaluation submodule, and associated for according to the time, the skip rate of constituent parts time and described programme content being carried out, determine that the resource content evaluation information of described video is as quality evaluation information.
In optional embodiment of the present invention, also comprise: user's sort module, for obtaining user profile, and classified according to described user's information; The quality determination module also comprises: the user estimates submodule, the programme content of lower the user skip rate in the time and association thereof at constituent parts for each classification of statistics respectively, determine user's evaluation information of described video, described user's evaluation information is added in described quality evaluation information.
Wherein, the quality determination module can comprise that mass value evaluation submodule, resource content evaluation submodule and user estimate at least one submodule in submodule.
In sum, by all types of operation behaviors and temporal information thereof to the user, added up, can obtain the operation behavior information in the constituent parts time, the type of operation behavior comprises following at least one: F.F., rewind down, drag, stop to play and close player, the type of operation behavior is various, can improve the accuracy of data.And operation behavior can be definite according to the weight of dissimilar correspondence weighted value, and then the skip rate of unit of account time, make the more accurate of skip rate.
Secondly, the quality evaluation information of video can be based on the mass value of skip rate, is a quantized value that standard is unified.Also can be further associated by the skip rate of unit interval and programme content, obtain content-based quality evaluation information, can also determine skip rate and associated programme content thereof that the user classified in the lower unit interval, obtain the quality evaluation information based on the user, thereby make quality evaluation information more comprehensive.
For device embodiment, because it is substantially similar to embodiment of the method, so description is fairly simple, relevant part gets final product referring to the part explanation of embodiment of the method.
Each embodiment in this specification all adopts the mode of going forward one by one to describe, and what each embodiment stressed is and the difference of other embodiment that between each embodiment, identical similar part is mutually referring to getting final product.
The present invention can describe in the general context of the computer executable instructions of being carried out by computer, for example program module.Usually, program module comprises the routine carrying out particular task or realize particular abstract data type, program, object, assembly, data structure etc.Also can in distributed computing environment (DCE), put into practice the present invention, in these distributed computing environment (DCE), be executed the task by the teleprocessing equipment be connected by communication network.In distributed computing environment (DCE), program module can be arranged in the local and remote computer-readable storage medium that comprises memory device.
Finally, also it should be noted that, in this article, relational terms such as the first and second grades only is used for an entity or operation are separated with another entity or operating space, and not necessarily requires or imply between these entities or operation the relation of any this reality or sequentially of existing.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thereby make the process, method, commodity or the equipment that comprise a series of key elements not only comprise those key elements, but also comprise other key elements of clearly not listing, or also be included as the intrinsic key element of this process, method, commodity or equipment.In the situation that not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, commodity or the equipment that comprises described key element and also have other identical element.
The above mass analysis method to a kind of video provided by the present invention and device, be described in detail, applied specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment is just for helping to understand method of the present invention and core concept thereof; , for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention simultaneously.
Claims (14)
1. the mass analysis method of a video, is characterized in that, comprising:
In video display process, detect respectively constituent parts operation behavior information time in of each user at video;
For each unit interval, by the weighted calculation of the operation behavior information to each user, determine the skip rate of video in the described unit interval;
Determine the quality evaluation information of described video in the skip rate of constituent parts time according to described video.
2. method according to claim 1, is characterized in that, in described video display process, detects respectively constituent parts operation behavior information time in of each user at video, comprising:
In video display process, detect each user's operation behavior and the temporal information of described operation behavior;
Described video was divided according to the unit interval, and the operation behavior by constituent parts in the time and the temporal information of described operation behavior form the operation behavior information of user within the described unit interval.
3. method according to claim 2, is characterized in that, in described video display process, detects each user's operation behavior and the temporal information of described operation behavior, comprising:
In video display process, detect each user's all types of operation behavior, wherein, the type of described operation behavior comprises following at least one: F.F., rewind down, drag, stop to play and close player;
Record the temporal information of the time of origin of operations behavior as described operation behavior;
And when the type of described operation behavior is F.F., rewind down or while dragging, the concluding time of recording described operation behavior, and the described concluding time is added in the temporal information of described operation behavior.
4. method according to claim 1, is characterized in that, described for each unit interval, by the weighted calculation of the operation behavior information to each user, determines that video, in the skip rate of described unit interval, comprising:
Within each unit interval of video, the operation behavior information to each user within the described unit interval is added up;
According to preset weight parameter, operation behavior information is weighted to calculating, determines the skip rate of video in the described unit interval.
5. method according to claim 1, is characterized in that, according to described video, determines the quality evaluation information of described video in the skip rate of constituent parts time, comprising:
Described video is compared in the skip rate of constituent parts time and the preset thresholding of skipping, and determine that according to comparative result the play quality value is as quality evaluation information.
6. method according to claim 1, is characterized in that, also comprises:
Described video is divided according to programme content, and determined the reproduction time of each section programme content;
According to described video, determine the quality evaluation information of described video in the skip rate of constituent parts time, comprise: according to the time, the skip rate of constituent parts time and described programme content are carried out associatedly, determine that the resource content evaluation information of described video is as quality evaluation information.
7. method according to claim 6, is characterized in that, also comprises:
Obtain user profile, and classified according to described user's information;
The programme content of the lower user of each classification of the statistics skip rate in the time and association thereof at constituent parts respectively, determine and user's evaluation information of described video to described user's evaluation information added in described quality evaluation information.
8. the quality evaluation device of a video, is characterized in that, comprising:
The behavior detection module, at video display process, detect respectively constituent parts operation behavior information time in of each user at video;
The skip rate determination module, for for each unit interval, by the weighted calculation of the operation behavior information to each user, determine the skip rate of video in the described unit interval;
The quality determination module, for determining the quality evaluation information of described video according to described video in the skip rate of constituent parts time.
9. device according to claim 8, is characterized in that, described behavior detection module comprises:
Detection sub-module, at video display process, detect each user's operation behavior and the temporal information of described operation behavior;
Divide and form submodule, for described video was divided according to the unit interval, the operation behavior by constituent parts in the time and the temporal information of described operation behavior form the operation behavior information of user within the described unit interval.
10. device according to claim 9, is characterized in that, described detection sub-module comprises:
The operation behavior detecting unit, at video display process, detect each user's all types of operation behavior, and wherein, the type of described operation behavior comprises following at least one: F.F., rewind down, drag, stop to play and close player;
Record cell, the temporal information for the time of origin that records the operations behavior as described operation behavior; And when the type of described operation behavior is F.F., rewind down or while dragging, the concluding time of recording described operation behavior, and the described concluding time is added in the temporal information of described operation behavior.
11. device according to claim 8, is characterized in that, described skip rate determination module comprises:
The statistics submodule, within each unit interval of video, the operation behavior information to each user within the described unit interval is added up;
Calculating sub module, be weighted calculating for the weight parameter according to preset to operation behavior information, determines the skip rate of video in the described unit interval.
12. device according to claim 8, is characterized in that, described quality determination module comprises:
Mass value is estimated submodule, for described video is compared in the skip rate of constituent parts time and the preset thresholding of skipping, and determines that according to comparative result the play quality value is as quality evaluation information.
13. method according to claim 8, is characterized in that, also comprises:
The division of teaching contents module, for described video is divided according to programme content, and the reproduction time of definite each section programme content;
The quality determination module comprises: the resource content evaluation submodule, and associated for according to the time, the skip rate of constituent parts time and described programme content being carried out, determine that the resource content evaluation information of described video is as quality evaluation information.
14. device according to claim 13, is characterized in that, also comprises:
User's sort module, for obtaining user profile, and classified according to described user's information;
The quality determination module also comprises: the user estimates submodule, the programme content of lower the user skip rate in the time and association thereof at constituent parts for each classification of statistics respectively, determine user's evaluation information of described video, described user's evaluation information is added in described quality evaluation information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310425590.8A CN103501434A (en) | 2013-09-17 | 2013-09-17 | Method and device for analyzing quality of video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310425590.8A CN103501434A (en) | 2013-09-17 | 2013-09-17 | Method and device for analyzing quality of video |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103501434A true CN103501434A (en) | 2014-01-08 |
Family
ID=49866597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310425590.8A Pending CN103501434A (en) | 2013-09-17 | 2013-09-17 | Method and device for analyzing quality of video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103501434A (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104168506A (en) * | 2014-08-13 | 2014-11-26 | 广州市香港科大霍英东研究院 | Method for obtaining user video experience quality by sniffing HTTP data packet |
| CN104410877A (en) * | 2014-11-28 | 2015-03-11 | 北京国双科技有限公司 | Method and device for processing user behavior data of network television |
| CN104486649A (en) * | 2014-12-18 | 2015-04-01 | 北京百度网讯科技有限公司 | Video content rating method and device |
| CN104519406A (en) * | 2014-11-13 | 2015-04-15 | 北京国双科技有限公司 | Method and device for monitoring network video clip playing behaviors |
| CN104581400A (en) * | 2015-02-10 | 2015-04-29 | 飞狐信息技术（天津）有限公司 | Video content processing method and video content processing device |
| CN104918067A (en) * | 2014-03-12 | 2015-09-16 | 乐视网信息技术（北京）股份有限公司 | Method and system for performing curve processing on video hot degree |
| WO2016062220A1 (en) * | 2014-10-20 | 2016-04-28 | 北京国双科技有限公司 | Video playing detection method and device |
| CN106131608A (en) * | 2016-07-29 | 2016-11-16 | 王晓光 | The viewing statistical method of different types of video file and system |
| CN106303675A (en) * | 2016-08-24 | 2017-01-04 | 北京奇艺世纪科技有限公司 | A kind of video segment extracting method and device |
| CN106341683A (en) * | 2016-08-24 | 2017-01-18 | 乐视控股（北京）有限公司 | Panoramic video quality judgment method and panoramic video quality judgment system |
| CN107094269A (en) * | 2016-11-04 | 2017-08-25 | 口碑控股有限公司 | Data processing method, equipment and video play device |
| CN107147957A (en) * | 2017-04-19 | 2017-09-08 | 北京小米移动软件有限公司 | Video broadcasting method and device |
| WO2018018594A1 (en) * | 2016-07-29 | 2018-02-01 | 王晓光 | Method and system for collecting statistics about view of video files of different types |
| CN107786895A (en) * | 2017-10-18 | 2018-03-09 | 北京奇艺世纪科技有限公司 | A kind of method for evaluating quality and device of broadcast page video recommendations |
| WO2018098884A1 (en) * | 2016-11-29 | 2018-06-07 | 深圳Tcl新技术有限公司 | Network video playback information acquisition method and system of mart television |
| CN109168044A (en) * | 2018-10-11 | 2019-01-08 | 北京奇艺世纪科技有限公司 | A kind of determination method and device of video features |
| CN109448548A (en) * | 2018-12-13 | 2019-03-08 | 沈峰 | A kind of management mode of indoor ad content |
| WO2019144838A1 (en) * | 2018-01-24 | 2019-08-01 | 北京一览科技有限公司 | Method and apparatus for use in acquiring evaluation result information of video |
| CN110267103A (en) * | 2019-05-13 | 2019-09-20 | 深圳市亿联智能有限公司 | Multimedia points-scoring system and method based on remote controler feedback |
| CN110392279A (en) * | 2018-04-17 | 2019-10-29 | 腾讯科技（深圳）有限公司 | Video broadcasting method, relevant device and computer readable storage medium |
| CN112104912A (en) * | 2020-11-20 | 2020-12-18 | 深圳市房多多网络科技有限公司 | Method, device and equipment for detecting video watching completion degree and readable storage medium |
| CN112565741A (en) * | 2020-11-26 | 2021-03-26 | 北京达佳互联信息技术有限公司 | Video fault processing method and device, equipment and storage medium |
| CN114598934A (en) * | 2022-02-11 | 2022-06-07 | 山东悦知教育科技有限公司 | Education software data processing method based on big data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101448174A (en) * | 2008-12-26 | 2009-06-03 | 深圳华为通信技术有限公司 | Image quality evaluation device and method thereof |
| CN102265609A (en) * | 2008-12-26 | 2011-11-30 | 富士通株式会社 | Program data processing device, method, and program |
| CN102487456A (en) * | 2009-11-30 | 2012-06-06 | 国际商业机器公司 | Method for providing visit rate of online video and device thereof |
-
2013
- 2013-09-17 CN CN201310425590.8A patent/CN103501434A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101448174A (en) * | 2008-12-26 | 2009-06-03 | 深圳华为通信技术有限公司 | Image quality evaluation device and method thereof |
| CN102265609A (en) * | 2008-12-26 | 2011-11-30 | 富士通株式会社 | Program data processing device, method, and program |
| CN102487456A (en) * | 2009-11-30 | 2012-06-06 | 国际商业机器公司 | Method for providing visit rate of online video and device thereof |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104918067A (en) * | 2014-03-12 | 2015-09-16 | 乐视网信息技术（北京）股份有限公司 | Method and system for performing curve processing on video hot degree |
| CN104168506A (en) * | 2014-08-13 | 2014-11-26 | 广州市香港科大霍英东研究院 | Method for obtaining user video experience quality by sniffing HTTP data packet |
| WO2016062220A1 (en) * | 2014-10-20 | 2016-04-28 | 北京国双科技有限公司 | Video playing detection method and device |
| US10104428B2 (en) | 2014-10-20 | 2018-10-16 | Beijing Gridsum Technology Co., Ltd. | Video playing detection method and apparatus |
| CN104519406B (en) * | 2014-11-13 | 2017-10-27 | 北京国双科技有限公司 | Internet video fragment plays the monitoring method and device of behavior |
| CN104519406A (en) * | 2014-11-13 | 2015-04-15 | 北京国双科技有限公司 | Method and device for monitoring network video clip playing behaviors |
| CN104410877A (en) * | 2014-11-28 | 2015-03-11 | 北京国双科技有限公司 | Method and device for processing user behavior data of network television |
| CN104486649A (en) * | 2014-12-18 | 2015-04-01 | 北京百度网讯科技有限公司 | Video content rating method and device |
| CN104486649B (en) * | 2014-12-18 | 2018-05-04 | 北京百度网讯科技有限公司 | Video content ranking method and device |
| CN104581400A (en) * | 2015-02-10 | 2015-04-29 | 飞狐信息技术（天津）有限公司 | Video content processing method and video content processing device |
| WO2018018594A1 (en) * | 2016-07-29 | 2018-02-01 | 王晓光 | Method and system for collecting statistics about view of video files of different types |
| CN106131608A (en) * | 2016-07-29 | 2016-11-16 | 王晓光 | The viewing statistical method of different types of video file and system |
| CN106303675B (en) * | 2016-08-24 | 2019-11-15 | 北京奇艺世纪科技有限公司 | A kind of video clip extracting method and device |
| CN106341683A (en) * | 2016-08-24 | 2017-01-18 | 乐视控股（北京）有限公司 | Panoramic video quality judgment method and panoramic video quality judgment system |
| CN106303675A (en) * | 2016-08-24 | 2017-01-04 | 北京奇艺世纪科技有限公司 | A kind of video segment extracting method and device |
| CN107094269A (en) * | 2016-11-04 | 2017-08-25 | 口碑控股有限公司 | Data processing method, equipment and video play device |
| WO2018098884A1 (en) * | 2016-11-29 | 2018-06-07 | 深圳Tcl新技术有限公司 | Network video playback information acquisition method and system of mart television |
| CN107147957B (en) * | 2017-04-19 | 2019-09-10 | 北京小米移动软件有限公司 | Video broadcasting method and device |
| CN107147957A (en) * | 2017-04-19 | 2017-09-08 | 北京小米移动软件有限公司 | Video broadcasting method and device |
| CN107786895B (en) * | 2017-10-18 | 2019-09-17 | 北京奇艺世纪科技有限公司 | A kind of method for evaluating quality and device of broadcast page video recommendations |
| CN107786895A (en) * | 2017-10-18 | 2018-03-09 | 北京奇艺世纪科技有限公司 | A kind of method for evaluating quality and device of broadcast page video recommendations |
| WO2019144838A1 (en) * | 2018-01-24 | 2019-08-01 | 北京一览科技有限公司 | Method and apparatus for use in acquiring evaluation result information of video |
| CN110392279A (en) * | 2018-04-17 | 2019-10-29 | 腾讯科技（深圳）有限公司 | Video broadcasting method, relevant device and computer readable storage medium |
| CN109168044B (en) * | 2018-10-11 | 2021-07-23 | 北京奇艺世纪科技有限公司 | Method and device for determining video characteristics |
| CN109168044A (en) * | 2018-10-11 | 2019-01-08 | 北京奇艺世纪科技有限公司 | A kind of determination method and device of video features |
| CN109448548A (en) * | 2018-12-13 | 2019-03-08 | 沈峰 | A kind of management mode of indoor ad content |
| CN110267103A (en) * | 2019-05-13 | 2019-09-20 | 深圳市亿联智能有限公司 | Multimedia points-scoring system and method based on remote controler feedback |
| CN112104912B (en) * | 2020-11-20 | 2021-02-12 | 深圳市房多多网络科技有限公司 | Method, device and equipment for detecting video watching completion degree and readable storage medium |
| CN112104912A (en) * | 2020-11-20 | 2020-12-18 | 深圳市房多多网络科技有限公司 | Method, device and equipment for detecting video watching completion degree and readable storage medium |
| CN112565741A (en) * | 2020-11-26 | 2021-03-26 | 北京达佳互联信息技术有限公司 | Video fault processing method and device, equipment and storage medium |
| CN112565741B (en) * | 2020-11-26 | 2023-04-28 | 北京达佳互联信息技术有限公司 | Video fault processing method and device, equipment and storage medium |
| CN114598934A (en) * | 2022-02-11 | 2022-06-07 | 山东悦知教育科技有限公司 | Education software data processing method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103501434A (en) | Method and device for analyzing quality of video | |
| CN103501461A (en) | Method for playing video and player | |
| US10362353B2 (en) | Video advertisement overlay system and method | |
| US10652605B2 (en) | Visual hot watch spots in content item playback | |
| US9715901B1 (en) | Video preview generation | |
| US9736432B2 (en) | Identifying popular network video segments | |
| Napoli | Audience evolution: New technologies and the transformation of media audiences | |
| CN104079997A (en) | Recommending system for digital television personalized programs and method thereof | |
| CN110225361B (en) | Live broadcast room recommendation method, storage medium, electronic device and system | |
| US20130232516A1 (en) | Method And Apparatus for Collection and Analysis of Real-Time Audience Feedback | |
| CN107710199A (en) | System and method for providing content in contents list | |
| CN104881237A (en) | Internet interaction method and client | |
| CN107846629A (en) | Recommend the method, apparatus and server of video to user | |
| CN107197368A (en) | Determine method and system of the user to multimedia content degree of concern | |
| US20150296195A1 (en) | Displaying content between loops of a looping media item | |
| CN104967690B (en) | A kind of information-pushing method and device | |
| CN102495840A (en) | Search engine for video website | |
| WO2022247220A9 (en) | Interface processing method and apparatus | |
| CN106802913A (en) | One kind plays content recommendation method and its device | |
| CN108668147A (en) | A kind of method and mobile terminal of optimization live streaming application | |
| CN104581224B (en) | Switch the method, apparatus and terminal of broadcasting content | |
| CN112218126A (en) | Content recommendation effect feedback method, system and equipment based on terminal | |
| CN105100840A (en) | Method and device for inter-cut of recommended media information | |
| CN104581226A (en) | Method and device for recommending programs by aiming at current live stream | |
| US8751527B1 (en) | Information retrieval system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140108 |
|
| RJ01 | Rejection of invention patent application after publication |