CN102833625B - The device and method of dynamic product placement in video - Google Patents
The device and method of dynamic product placement in video Download PDFInfo
- Publication number
- CN102833625B CN102833625B CN201210298196.8A CN201210298196A CN102833625B CN 102833625 B CN102833625 B CN 102833625B CN 201210298196 A CN201210298196 A CN 201210298196A CN 102833625 B CN102833625 B CN 102833625B
- Authority
- CN
- China
- Prior art keywords
- trade mark
- image
- video
- picture
- module
- 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.)
- Expired - Fee Related
Links
Landscapes
- Editing Of Facsimile Originals (AREA)
Abstract
The invention discloses a kind of device and method of dynamic product placement in video, this device mainly comprises: for original video being resolved into the video decomposing module of the image of single frames; For utilizing the picture recognition module of image recognition and the trade mark edited in image described in algorithm identified or object; For the trade mark replacement module of replacing the trade mark in described image; And for the Video Composition module of the new video of the described Images uniting by transposed trade mark.Adopt the present invention, the ad content dynamically implanting custom formats in video can be realized, increase correlation and the effect of advertisement.
Description
Technical field
The present invention relates to the Internet, television advertising play-back technology, particularly relate to a kind of device and method of dynamic product placement in video.
Background technology
TV and advertisement are complementary always, and advertisement broadcast mode that can be present is but very stiff.Introduce " soft advertisement " in more existing video or television programs now, such as, the stage property that the essay of Spring Festival Gala uses is with the printed words of Sohu, and this is the one innovation in brand advertising field.But the advertisement of this implantation is static, can not change as required.If, company of Sina is as the competitor of Sohu.com Inc., when the website of oneself being broadcasted like this with the video frequency program of rival as information such as the LOGO of " Sohu " or registered trade marks, be do not wish that the information of rival appears in video frequency program.A lot of advertisers also wish can based on context with the environment residing for video to replace trade mark.Such as, the trade mark of Coca-Cola has varicolored, also wishes the trade mark replacing different colours, increase the identifiability of trade mark, thus strengthen the effect of advertisement under different backgrounds.
Summary of the invention
In view of this, main purpose of the present invention is to provide a kind of device and method of dynamic product placement in video, by dynamically implanting or changing the contents such as brand LOGO in video, to reach TV programme or the contextual needs of web site contents.
For achieving the above object, technical scheme of the present invention is achieved in that
A device for dynamic product placement, mainly comprises video decomposing module, picture recognition module, trade mark replacement module and Video Composition module in video; Wherein:
Video decomposing module, for resolving into the image of single frames by original video;
Picture recognition module, for utilizing image recognition technology and the trade mark in image described in editor's algorithm identified or object;
Trade mark replacement module, for replacing the trade mark in described image; And
Video Composition module, for the video that the described Images uniting by transposed trade mark is new.
Wherein, described video decomposing module, also for described single-frame images to be resolved into further one or more pictures of lower pixel according to the resolution of image self and size.
Picture recognition module, also for carrying out fingerprint calculating to the low pixel picture after each decomposition according to the information of each pixel, and mates with the fingerprint of the segmentation figure comprising trademark information.
A method for dynamic product placement in video, the method comprises:
A, original video is resolved into the step of the image of single frames;
B, utilize the step of trade mark described in image recognition technology and editor's algorithm identified in image or object;
C, the step that the trade mark in described image is replaced;
D, by the step of video new for the described Images uniting of transposed trade mark.
Wherein, steps A also comprises: the image after being decomposed by every frame resolves into one or more pictures of lower pixel further according to the resolution of image self and size.
Utilize the process of image recognition technology and the trade mark edited in image described in algorithm identified or object described in step B, comprising:
B1, to described each again decompose after picture, perform fingerprint algorithm according to the Pixel Information of image, and the image after decomposing mated with the little figure fingerprint at trade mark place to be replaced;
If the match is successful for B2, then according to the adjacent principle of equity, adjacent picture is also compared; If namely the adjacent picture of the little figure of the decomposition of Target Photo and the adjacent picture of the little figure of decomposition of the trade mark picture that the match is successful match, finger print information is consistent, then matching result consistent with the finger print information of Target Photo (can increase the accuracy of coupling), then think and have found Target Photo.
Described in step C to the process that the trade mark in described image is replaced be:
Utilize the image feature information according to the target figure found, find candidate's pictorial trademark of coupling, then Target Photo is replaced as the candidate's trade mark matched.
Described in step C to the process that the trade mark in described image is replaced be:
Original trademark is removed from every frame picture of original video, and selects suitable trade mark to replace according to its background color.
Described in step C to the process that the trade mark in described image is replaced be: the displacement carrying out trade mark according to the context of webpage or the demand of advertisement.
The device and method of dynamic product placement in video provided by the present invention, has the following advantages:
The present invention utilizes image recognition technology and editor's algorithm, realizes the ad content (comprising image, word etc.) dynamically implanting custom formats in video.By this editor's algorithm, the image after displacement is merged into new video automatically, still can, nature equally true with the picture of former video.Utilize technology of the present invention, according to the context of the page residing for video, can also comprise network address, the word, keyword etc. on video embedded location side, carry out the displacement of advertisement, thus add correlation and the effect of advertisement.
Accompanying drawing explanation
Fig. 1 is the structural representation of the present invention's dynamic device of product placement in video;
Fig. 2 is the method flow diagram of the present invention's dynamic product placement in video.
Embodiment
Below in conjunction with accompanying drawing and embodiments of the invention, device and method of the present invention is described in further detail.
Fig. 1 is the structural representation of the present invention's dynamic device of product placement in video, and as shown in Figure 1, this device mainly comprises video decomposing module 10, picture recognition module 11, trade mark replacement module 12, Video Composition module 13.Wherein:
Described video decomposing module 10, for resolving into the image of single frames by original video.
Utilize video decomposing module 10, one section of video (original video) is decomposed into image one by one, then every two field picture is resolved into little figure further.The size of the image after decomposing is determined according to the resolution of image self and size.Such as: be the little figure that the picture breakdown of 1024 × 720 pixels becomes 8 × 8 pixels by resolution.
Picture recognition module 11, for utilizing image recognition technology and the trade mark in image described in editor's algorithm identified or object.
Utilize this picture recognition module 11, little figure after each is decomposed, Pixel Information according to image after each decomposition performs fingerprint (Fingerprinting) algorithm, and the image after decomposing is mated with the little figure fingerprint at trade mark place to be replaced, if the match is successful, then adjacent picture is also compared.It should be noted that matching result must consistent with Target Photo, judge to think and have found Target Photo.
Trade mark replacement module 12, for replacing the trade mark in described image.
Here, utilizing the characteristics of image of target figure according to finding, as the information such as color of background, domain color, adjacent picture, finding candidate's pictorial trademark of comparison match, then Target Photo being replaced as the candidate's trade mark matched.When carrying out trade mark displacement, might not only replace the little figure of decomposition.Also can cut according to the profile of target trade mark or border, and in line of cut, be replaced as candidate's trade mark.In order to make the trade mark after displacement and the image near it well merge, Fuzzy Processing can be carried out to the border of segmentation, seeming more natural to make the trade mark after replacing.
Video Composition module 13, for being reconfigured by the described Target Photo of transposed trade mark, forms new continuous print video.
As the solution that another kind is parallel: in described picture recognition module 11, other image recognition technology can also be used.Such as, first picture is carried out to the process of profile, mate with the profile of target trade mark in the profile identified, then the image transform in profile is become target trade mark.This trade mark often appearing at some regular contours (as circle, cylindricality, the feature such as square) inner for some is more applicable.Such as, the trade mark of Cocacola generally appears on pop can, so, identify on pop can containing trade mark image after, can whole pop can image change into Pepsi mark.
In fact, when identifying target trade mark, which trade mark we might not specify.When implementing of the present invention, if set up the image data base of the trade mark that is enriched very much, can decompose every two field picture and mate with all images of this image data base, and find one or more trade mark.Set up a database to candidate's icon, this database not only can have image information, also has text description, and other attributes, as classification, price etc., then according to gray requirement, or according to the context of video, displacement trade mark is carried out to the trademark image in video.Here, described context, can also comprise the search keyword in search engine, after replacing, shows the result after trade mark displacement in Search Results.
Fig. 2 is the method flow diagram of the present invention's dynamic product placement in video, and as shown in Figure 2, the method mainly comprises the steps:
Step 20: image original video being resolved into single frames.First target trade mark and candidate's trade mark are produced to the image of different resolution, then each picture breakdown Cheng little Tu, calculate the fingerprint (Fingerprint) of each little figure.Corresponding fingerprint algorithm can with reference to existing algorithm, and e.g., Radon converts (Radon Transformation).
Step 21: utilize the trade mark in image described in image recognition and editor's algorithm identified or object.Utilize Video Analysis Technology, analyze the trade mark or object that comprise in described video.This process can be automatically find these trade marks, also can be some trade marks of specifying.Can according to the characteristics of image of Target Photo, the color etc. as background, domain color, adjacent picture carries out the candidate's pictorial trademark analyzing, find comparison match.
Step 22: the trade mark in described image is replaced.
Here, according to the characteristics of image of the above-mentioned Target Photo found, as the color etc. of background, domain color, adjacent picture, after finding candidate's pictorial trademark of comparison match, described Target Photo is replaced as the candidate's trade mark matched.We according to the demand of reality, as the requirement according to advertiser, can also replace these trade marks.
In fact, our algorithm can remove such trade mark (editor by every frame picture) from original video, and according to its background color, selects suitable trade mark to replace.Such as, the trade mark of the meeting configuration white of red background.These matched rules, can be specified by advertiser, or use our matching algorithm, thus allow the trade mark of new displacement show very natural in video.
In addition, the displacement of trade mark can also be carried out according to the context of webpage.Such as, when webpage is when introducing tourist attractions, Pekinese, the trade mark in our video just can be replaced as the relevant native article in Beijing, thus reaches better advertising results.
Step 23: last, by video new for the described Images uniting of transposed trade mark.
The present invention can also be applied in the implantation of new advertisement.Advertiser can define and will occur trade mark on the object of which class, and then, our image recognition technology can identify this type objects, and shows trade mark above.Such as, occur a camera lens drunk water with beverage bottle in video, we will be replaced as advertiser's the trade mark of this beverage bottle dynamically.Again such as, the billboard in competition gymnasiums and stadiums also can be replaced as the advertisement mode that advertiser requires.In addition, the present invention also can be used for the editor of Online Video, is also applicable to the video editing to TV programme.
The above, be only preferred embodiment of the present invention, be not intended to limit protection scope of the present invention.
Claims (2)
1. a device for dynamic product placement in video, is characterized in that, mainly comprise video decomposing module, picture recognition module, trade mark replacement module and Video Composition module; Wherein:
Video decomposing module, for original video being resolved into the image of single frames, also for described single-frame images to be resolved into further one or more pictures of lower pixel according to the resolution of image self and size;
Picture recognition module, for utilizing image recognition technology and the trade mark in image described in editor's algorithm identified or object, also for carrying out fingerprint calculating to the low pixel picture after each decomposition according to the information of each pixel, and mate with the fingerprint comprising the segmentation figure of trademark information at trade mark place to be replaced;
Trade mark replacement module, for replacing the trade mark in described image; And
Video Composition module, for the video that the described Images uniting by transposed trade mark is new.
2. a method for dynamic product placement in video, it is characterized in that, the method comprises:
A, original video is resolved into the step of the image of single frames; Image after being decomposed by every frame resolves into one or more pictures of lower pixel further according to the resolution of image self and size;
B, utilize the step of trade mark described in image recognition technology and editor's algorithm identified in image or object; The described process utilizing image recognition technology and the trade mark edited in image described in algorithm identified or object, comprise: B1, to each again decompose after picture, Pixel Information according to image performs fingerprint algorithm, and is mated with the little figure fingerprint at trade mark place to be replaced by the image after decomposing; If the match is successful for B2, then according to the adjacent principle of equity, adjacent picture is also compared; If namely the adjacent picture of the little figure of the decomposition of Target Photo and the adjacent picture of the little figure of decomposition of the trade mark picture that the match is successful match, finger print information is consistent, then think and have found Target Photo;
C, the step that the trade mark in described image is replaced; Describedly to the process that the trade mark in described image is replaced be:
Utilize the image feature information according to the target figure found, find candidate's pictorial trademark of coupling, then Target Photo is replaced as the candidate's trade mark matched; Or, original trademark is removed from every frame picture of original video, and selects suitable trade mark to replace according to its background color; Or the displacement of trade mark is carried out according to the context of webpage or gray demand;
D, by the step of video new for the described Images uniting of transposed trade mark.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210298196.8A CN102833625B (en) | 2012-08-21 | 2012-08-21 | The device and method of dynamic product placement in video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210298196.8A CN102833625B (en) | 2012-08-21 | 2012-08-21 | The device and method of dynamic product placement in video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102833625A CN102833625A (en) | 2012-12-19 |
| CN102833625B true CN102833625B (en) | 2015-10-28 |
Family
ID=47336547
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210298196.8A Expired - Fee Related CN102833625B (en) | 2012-08-21 | 2012-08-21 | The device and method of dynamic product placement in video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102833625B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103024480B (en) * | 2012-12-28 | 2016-06-01 | 杭州泰一指尚科技有限公司 | A kind of method embedding advertisement in video |
| US20160300270A1 (en) * | 2013-11-21 | 2016-10-13 | Huawei Technologies Co., Ltd. | Method and Apparatus for Embedding Advertisement into File, System and Terminal |
| WO2015100518A1 (en) | 2013-12-31 | 2015-07-09 | Google Inc. | Systems and methods for converting static image online content to dynamic online content |
| CN103974126B (en) * | 2014-05-15 | 2017-03-01 | 北京奇艺世纪科技有限公司 | A kind of method and device of product placement in video |
| CN104363485A (en) * | 2014-11-28 | 2015-02-18 | 北京奇艺世纪科技有限公司 | Video-label-based method and system for delivering information |
| CN104618745B (en) * | 2015-02-17 | 2017-08-01 | 北京影谱科技股份有限公司 | A device for dynamically embedding advertisements in videos |
| CN104735465B (en) * | 2015-03-31 | 2019-04-12 | 北京奇艺世纪科技有限公司 | The method and device of plane pattern advertisement is implanted into video pictures |
| CN106375858B (en) * | 2015-07-23 | 2019-06-18 | Tcl集团股份有限公司 | A kind of method, apparatus and system of product placement in video |
| CN105141987B (en) * | 2015-08-14 | 2019-04-05 | 京东方科技集团股份有限公司 | Advertisement method for implantation and advertisement implant system |
| CN105338404B (en) * | 2015-10-29 | 2019-03-29 | 北京击壤科技有限公司 | A kind of TV programme are implanted into soft wide identifying system and its method |
| CN110662111A (en) * | 2018-06-29 | 2020-01-07 | 南京芝兰人工智能技术研究院有限公司 | Method and system for implanting content information in video in batch |
| CN111126372B (en) * | 2019-11-29 | 2023-07-11 | 北京金山云网络技术有限公司 | Marking method, device and electronic equipment for logo area in video |
| CN110992251B (en) * | 2019-11-29 | 2023-11-10 | 北京金山云网络技术有限公司 | Logo replacing method and device in video and electronic equipment |
| CN113298926A (en) * | 2020-02-06 | 2021-08-24 | 阿里巴巴集团控股有限公司 | Video processing method, video processing device and electronic equipment |
| CN113055740A (en) * | 2020-04-26 | 2021-06-29 | 杜达亮 | Method and system for identifying and pasting object in video |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1324185A (en) * | 2000-04-26 | 2001-11-28 | 朗迅科技公司 | Method and device for dynamically modifying digital video image |
| CN101398829A (en) * | 2007-09-30 | 2009-04-01 | 国际商业机器公司 | Method and apparatus for marking and modifying video, and video processing method and apparatus |
| CN102036106A (en) * | 2009-10-07 | 2011-04-27 | 国际商业机器公司 | Insertion method and system of content into media streams |
-
2012
- 2012-08-21 CN CN201210298196.8A patent/CN102833625B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1324185A (en) * | 2000-04-26 | 2001-11-28 | 朗迅科技公司 | Method and device for dynamically modifying digital video image |
| CN101398829A (en) * | 2007-09-30 | 2009-04-01 | 国际商业机器公司 | Method and apparatus for marking and modifying video, and video processing method and apparatus |
| CN102036106A (en) * | 2009-10-07 | 2011-04-27 | 国际商业机器公司 | Insertion method and system of content into media streams |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102833625A (en) | 2012-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102833625B (en) | The device and method of dynamic product placement in video | |
| CN109716327B (en) | Video capturing framework for visual search platform | |
| US8543924B2 (en) | Contextual-display advertisement | |
| TWI631514B (en) | Method and system for marking recognition based on mobile terminal | |
| US9916499B2 (en) | Method and system for linking printed objects with electronic content | |
| EP2587826A1 (en) | Extraction and association method and system for objects of interest in video | |
| CN105141987A (en) | Advertisement implanting method and advertisement implanting system | |
| CN101398829B (en) | Method and apparatus for marking and modifying video, and video processing method and apparatus | |
| US20160104212A1 (en) | Computer implemented system for managing advertisements and a method thereof | |
| CN113010711B (en) | Method and system for automatically generating movie poster based on deep learning | |
| US9177199B2 (en) | Semantic magazine pages | |
| US20190065911A1 (en) | Media Intelligence Automation System | |
| CN110232131B (en) | Creative material searching method and device based on creative tag | |
| US20200285671A1 (en) | Terminal and apparatus for providing search information based on color information | |
| AU2020202627A1 (en) | System and method for retrievable storage, detection and reproduction of information for advertising media and information media | |
| CN105159923A (en) | Video image based article extraction, query and purchasing method | |
| CN118761812B (en) | AI-based advertisement material data labeling method and system | |
| CN114090630A (en) | Commodity data integration method based on distributed microservice cluster | |
| KR20100125878A (en) | Advertising system and method using object image extraction | |
| Manovich et al. | Understanding scanlation: how to read one million fan-translated manga pages | |
| US20210390306A1 (en) | Ar system for enhanced book covers and related methods | |
| CN110968804A (en) | A museum electronic label publicity system based on intelligent Internet of things | |
| Wang et al. | TextAtlas5M: A Large-scale Dataset for Dense Text Image Generation | |
| CN110969750A (en) | A museum visit guidance system based on intelligent voice | |
| JP2023032461A (en) | Region attribute specification method, region attribute specification device, description matter detection method, description matter detection device and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20151028 Termination date: 20200821 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |