Embodiment
Below in conjunction with Figure of description the specific embodiment of the present invention is described.
A kind of visual media audient information feedback system is provided among the present invention, and it provides a kind of effective audient's information feedback mechanism for visual media, thereby the input effect of visual information is write down and feeds back.
Please referring to shown in Figure 1, it shows block diagram among the embodiment of a kind of visual media audient of the present invention information feedback system, and described visual media audient information feedback system comprises visual media 10, video acquisition device 20, video analysis device 30 and information storage apparatus 40.Wherein, described visual media 10 is used to carry visual information and watches for spectators; Described video acquisition device 20 is used to gather the spectators' that watch visual information video; Described video analysis device 30 is used for the video of gathering is analyzed the spectators that watch visual information to determine; Described information storage apparatus 40 is used to write down the spectators that watch visual information initially watches constantly and stops to watch constantly.
In the present invention, visual media 10 can be a sensu lato notion, it can be a video player, promotional plate, displaying sales counter in the exhibition or the goods carrier in the supermarket, we can say so long as the carrier of visual information can be called as visual media, accordingly, visual information here also can be a sensu lato notion, it can be the video information of playing on video player, show the show piece on the sales counter, goods on the goods carrier or the advertisement information on the promotional plate be we can say so long as can be called as visual information by the artificial information that can see that is loaded on the visual media.In one embodiment, described video player can be common cable TV or the radiotelevisor that has among the present huge numbers of families, also can be to be positioned in the building or the special television machine in the elevator, even can be mobile TV, described video information can be video ads or video TV program etc., such as video ads or TV series program of one section Coca-Cola.Usually, be positioned in the building or some advertisements of meeting round-robin broadcast in the special television machine in the elevator, and broadcast TV programme and advertisement that common wired or wireless TV can replace.Described promotional plate can stand in the both sides, street or post up on the buildings for masses to be watched, and described advertisement information generally is a billboard, the huge image of representing for certain clothes brand such as the huge image or the Zhou Jielun of a certain automobile etc.
In a specific embodiment, see also shown in Figure 3, described visual media 10 can have several, can form a far-ranging visual media system like this to form the propagation of large-scale visual information, and several visual medias can disperse to be arranged on different geographic position.In one 20 layers building, possibility all will be placed a special television for every layer, all place a special television in each elevator, for another example in supermarket, have tens of in different positions or up to a hundred shelf, in exhibition, has the dozens of exhibition position for another example in different positions.Described video acquisition device and visual media are supporting settings, described video acquisition device can be placed on the corresponding visual media or near, promptly a visual media is equipped with corresponding video acquisition device, such as all being equipped with corresponding video acquisition device for every TV, for each shelf all is equipped with corresponding video acquisition device, for each exhibition position is equipped with corresponding video acquisition device.
The visual angle of described video acquisition device 20 is the visual orientation towards visual media, the visual field covers the part or all of viewing area of visual media, its objective is that when spectators really watched visual information, described video acquisition device 20 can collect these spectators' facial image.The visual orientation here is meant that spectators can see the direction of visual information, the viewing area is meant that spectators can see the zone of visual information, with the televisor is example, visual orientation not only comprises the direction of the dead ahead of screen towards screen, also comprise screen left front, right front, go up the place ahead, inferoanterior direction towards screen, the viewing area is meant the zone that can see TV screen.In one embodiment, the video acquisition device that disposes for each visual media can be a camera, such as selecting for use the wide-angle imaging head to cover the whole viewing areas of visual media, selects for use common camera to cover the main viewing area of visual media part.In another embodiment, the video acquisition device that disposes for each visual media can be several cameras, described several cameras can be placed in the different position of corresponding visual media, can overlap in their visual field, the overlapping part in the visual field is the topmost zones of watching visual media of spectators, such as the zone, dead ahead of a TV screen, can there be several cameras to photograph the overlapping part in the visual field like this, this helps the analysis of follow-up video analysis device.Certainly, the visual field of several cameras can be not overlapping fully yet, and they can responsible separately zone separately so.
See also shown in Fig. 2 and 3, described visual media audient information feedback system can also comprise video frequency transmitter 50, it is used for the video of receiver, video harvester 20 collections and transmits it to video analysis device 30, video frequency transmitter 50 can pass through the internet, the LAN (Local Area Network) transmission of video is to video analysis device, so just can make video analysis device 30 and video acquisition device 20 and visual media 10 split settings, described information storage apparatus 40 is split settings with video acquisition device 20 and visual media 10, and the spectators' of each visual media of collecting video is carried out collective analysis handle and preserve, be convenient to collection and management like this information.Be set at respectively such as several cameras and special television machine in the different floors or elevator of building, and video analysis device and information storage apparatus can be arranged on control center or other position of building.Certainly, also can be equipped with corresponding video analysis device and information storage apparatus, the mode of other accommodation can also be arranged in addition, just not enumerate one by one here for every cover visual media and video acquisition device.
The video of 30 pairs of collections of described video analysis device carries out people's face and detects, detect people's face, the people of detected people's face correspondence is regarded as the spectators that watch visual information, accordingly, determine now to watch the initial of spectators of visual information to watch constantly according to going out of detected people's face, determine constantly to watch according to the disappearance of detected people's face visual information spectators stop to watch constantly.Can adopt any one method that discloses in the prior art about concrete method for detecting human face, such as the technical scheme that on 06 28th, 2006 disclosed application numbers is 200510135668.8 patented claim, it has disclosed the real-time detection and the method and system that continue to follow the tracks of of people's face in a kind of video sequence, the technical scheme that is the patented claim of 200610113423.X on 03 07th, 2007 disclosed application number for another example, it has disclosed the detection of a kind of colourful attitude people's face and tracing system and method.
Here simply introducing number of patent application is that a kind of colourful attitude people's face that 200610113423.X discloses detects and method for tracing, and it comprises:
(1) by people's face sample training, obtain the positive and half side-view detection model of people's face respectively, and definite AAM faceform;
(2) utilize the positive and half side-view detection model of described people's face, inputted video image is carried out people's face detect, determine whether there is people's face in the two field picture;
(3) if in certain two field picture, detect people's face, then in subsequent frame, follow the trail of and verify this people's face, comprise step:
(31) the people's face position in the tracking former frame image, the preliminary position of people's face in the acquisition present frame;
(32) with the preliminary position of described acquisition as initial value, utilize the colourity difference of present frame and former frame image, calculate the point-to-point speed of people's face;
(33) according to described point-to-point speed, estimate the Position Approximate of people's face in the present frame, and utilize positive surface model of described people's face and half side-view detection model, near this position, detect, to verify this people's face;
(34) if detect people's face near this position, then checking is passed through, and adopts described AAM faceform to calculate the affined transformation coefficient of working as forefathers' face, obtains the characteristic parameter of present frame people face.
In conjunction with the present invention, video analysis device 30 also needs to carry out people's face sample training in advance, obtains the positive and half side-view detection model of people's face respectively, and definite AAM faceform, concrete process can be identical with this documents, also can adopt existing other training tools.Then, in step 2) in video analysis device 30 utilize described people's face positive with the half side-view detection model, video acquisition device 20 inputted video image are carried out people's face to be detected, determine whether there is people's face in the two field picture, afterwards, if video analysis device 30 detects people's face in certain two field picture, then in subsequent frame, follow the trail of and verify this people's face, after checking is passed through, can determine that then video analysis device 30 has detected people's face, the people of detected people's face correspondence is regarded as the spectators that watch visual information, corresponding, can be with that two field picture time corresponding point the going out now after verifying as people's face.At last, proceed step 3) and continue this people's face is followed the tracks of, do not pass through, will not verify that two field picture time corresponding point of passing through as the disappearance of people's face constantly this moment up to checking.In a preferred embodiment, after detecting people's face and continuing the tracking verification scheduled duration, just spectators initial watched constantly and spectators stop to watch entry information save set 40 constantly, scheduled duration can be 2-5 second or other setting durations.Such as camera is that 24 frames/second beats is taken the photograph, and described video analysis device 30 detects and authenticates to after someone's face, just writes down current time and frame number, and for example 2007.1.1.08:45:21.14 represents the 14th frame in 8: 45: 21 morning of 1 month 1 day July in 2007.In each follow-up frame, this people's face is followed the tracks of, till it disappears, and the time when writing down this front face and disappearing, 2007.1.1.08:46:08.20 for example, like this, to everyone face, we can obtain its time of occurrence and die-out time, for example, someone's face: (2007.1.1.08:45:21.14,2007.1.1.08:46:08.20).
In a preferred embodiment, video analysis device 30 is after detecting people's face, needing that also detected people's face is carried out sight line judges, sight line is pointed to people's face of visual information and regarded as the spectators that watch visual information, so corresponding, point to according to sight line visual information people's face go out to determine now to watch the initial of spectators of visual information to watch constantly, point to the disappearance moment of people's face of visual information according to sight line, determine to watch visual information spectators stop to watch constantly, the result of Fen Xiing is more accurately like this.It is 200510077043.0 patented claim that concrete sight line decision method please refer to the application number that is disclosed on November 13rd, 2005, and it discloses a kind of method and apparatus of differentiating direction of visual lines, has just repeated no more here.
Described visual media 10 carries one or several visual informations successively, can all distribute a well-determined sign for each visual information, described information storage apparatus 40 is also preserved the sign of each visual information of visual media 10 its carryings and the initial moment and the finish time that this visual information occurs, such as 3 video ads of special television meeting loop play in the building, be respectively the Sprite advertisement, Cherry's automotive advertising and unified green tea advertisement, its sign can be respectively so: 01 represents the Sprite advertisement, 02 represents Cherry's automotive advertising, the unified green tea advertisement of 03 representative, can certainly use A, B, C comes when making a check mark, as long as this sign can uniquely be determined this then advertisement.In addition, also need to preserve the initial moment and the finish time that this visual information occurs, such as 01,2007.7.2.9:00:00-9:03:00,2007.7.2.9:30:30-9:33:30 ... this means that the Sprite advertisement began to play in o'clock sharp 2 days 9 July in 2007, play to whole end 9: 03 on the 2nd July in 2007, begin to play 9: 30: 30 on the 2nd July in 2007 again then, finish to play to 9: 33: 30 on the 2nd July in 2007.In certain embodiments, several visual medias 10 can be arranged, so can be for each visual media 10 all distribute a well-determined sign, the sign that described information storage apparatus 40 is preserved each visual media.Such as ten special televisions that have in: the building, their sign can be respectively so: 001 represents that of one deck, and 002 represents that of two layers, and 003 represents that of three layers ..., can certainly identify with other codes.Like this, the information of information storage apparatus record has just comprised the sign of each visual media, the sign of each visual information in each visual media and the initial moment and the finish time that this visual information occurs, whereby, we just can understand the visual information that arbitrary visual media carries at any time.
As previously mentioned, having write down the initial of spectators of watching visual information in the information storage apparatus 10 watches the moment and stops to watch the moment, because described video acquisition device 20 and visual media 10 are supporting settings, therefore the initial of described spectators watched the moment and stops to watch the moment at predetermined visual media.So far the information that described information storage apparatus is preserved can be organized into following formation: visual media identifies, the visual information sign, and the initial moment and the finish time that visual information occurs, the initial of every spectators watched the moment and stops to watch the moment.Information in the described information storage apparatus can be designated clue with visual media and put in order, and these information that this means visual media sign back are all corresponding to this visual media sign.In addition, information storage apparatus 10 can also be preserved the continuous sequence or the single image of people's face video image of the spectators that watch visual information.
By foregoing description as can be known, visual media audient information feedback system of the present invention is by disposing the video information that corresponding video acquisition device is collected spectators for visual media, we just can note audient's information like this, have realized the feedback of audient's information.Be example also with the special television in the building, we can find each video ads sign of playing on it by the sign of target-specific TV, the initial moment and the finish time that each video ads is play, every spectators of this target-specific TV initial watched constantly and stops to watch these information that constantly waits, and analyzes us and just can understand each then the watching person-time, tire out own viewing time etc. and can reflect the information of throwing in effect of video ads by own.
The information of preserving in the described information storage apparatus all is a little raw data, we can according to these raw data carry out some sorting-outs in statistics obtain some more intuitively, active data, and then more direct proof is thrown in effect.In a specific embodiment, see also shown in Figure 2, described visual media audient information feedback system also comprises the data statistics device, and it carries out data statistics according to the rule of setting to the data of preserving in the information storage apparatus and obtains the data that people relatively pay close attention to.Specifically, the data statistics device counts the watching person-time of predetermined visual information, accumulative total viewing time, average single viewing time, set and watch in the period person-time or set accumulative total viewing time in the period, and these data can be made correct decision-making for people and play very big help.In addition, we can go to set rule ourselves and go to count own needed data.
According to a further aspect in the invention, the present invention also provides a kind of visual media audient information approach, and it provides a kind of effective audient's information feedback mechanism for visual media, thereby the input effect of visual information is write down and feeds back.Because visual media audient information approach of the present invention and visual media audient massaging device are roughly the same on principle, for the preamble of getting along well repeats mutually, only are to briefly introduce visual media audient information approach of the present invention here.
Please referring to shown in Figure 4, it shows schematic flow sheet among the embodiment of a kind of visual media audient of the present invention information feedback method, described visual media is used to carry visual information to be watched for spectators, and it comprises: step 100, gather the spectators' that watch visual information video; Step 200 is analyzed the spectators that watch visual information to determine to the video of gathering; Step 300, record watch the initial of spectators of visual information to watch the moment and stop to watch the moment.
Described visual media can be a video player, and the visual information of its carrying is a video information; Described visual media can be a promotional plate, and the visual information of its carrying is a billboard; Described visual media can be the article display platform, and the visual information of its carrying is a displaying articles; Or described visual media can be shelf, and the visual information of its carrying is a goods.
The visual angle of the video acquisition in the step 100 is towards the visual orientation of visual media, and the visual field of video acquisition covers the part or all of viewing area of visual media.
In a specific embodiment, step 200 specifically comprises: b1) video of gathering is carried out people's face and detect; B2) people with detected people's face correspondence regards as the spectators that watch visual information; So then be to determine now to watch the initial of spectators of visual information to watch constantly in the step 300 according to going out of detected people's face, determine constantly to watch according to the disappearance of detected people's face visual information spectators stop to watch constantly.In a preferred embodiment, step 200 also comprises: b3) detected people's face is carried out sight line and judge, sight line is pointed to people's face of visual information and regarded as the spectators that watch visual information; So then in the step c) be according to sight line point to visual information people's face go out to determine now to watch the initial of spectators of visual information to watch constantly, the disappearance of pointing to people's face of visual information according to sight line determine constantly to watch visual information spectators stop to watch constantly.
Visual media carries one or several visual informations successively, for each visual information is all distributed a well-determined sign, also writes down the sign of each visual information of visual media carrying and the initial moment and the finish time that this visual information occurs in step 300.Visual media has several, for each visual media all distributes a well-determined sign, also writes down the sign of current visual media in step 300.In a preferred embodiment, described visual media audient information feedback method also comprises: go out to set the watching person-time of visual information, accumulative total viewing time, average single viewing time, set the step of watching in the period person-time or setting accumulative total viewing time in the period according to the Information Statistics of record in the step 300.
The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being done, is equal to replacement etc., all should be included within protection scope of the present invention.