CN103257992A - Method and system for retrieving similar videos - Google Patents
Method and system for retrieving similar videos Download PDFInfo
- Publication number
- CN103257992A CN103257992A CN2013100340117A CN201310034011A CN103257992A CN 103257992 A CN103257992 A CN 103257992A CN 2013100340117 A CN2013100340117 A CN 2013100340117A CN 201310034011 A CN201310034011 A CN 201310034011A CN 103257992 A CN103257992 A CN 103257992A
- Authority
- CN
- China
- Prior art keywords
- video
- frame
- eigenvector
- retrieved
- key frame
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a system for retrieving similar videos. The method includes: using a color space model to calculate feature vectors of each video key frame set in a video database; mapping the feature vectors through a hash function, and building indexes according to sections where mapped feature vectors are; determining number of a corresponding index of each section where each mapped feature vector of each motion vector of a to-be-retrieved video is, extracting video information corresponding to all feature vectors under corresponding indexes, and calculating similarity according to the to-be-retrieved video and similar video key frame number extracted from the corresponding indexes; and outputting videos corresponding to calculated results larger than threshold at retrieved results. By the method, efficiency and accuracy of similar video retrieving are increased.
Description
Technical field
The present invention relates to the Computer Applied Technology field, relate in particular to a kind of method and system of similar video retrieval.
Background technology
Along with the fast development of network flow-medium technology, increasing video and multi-medium data have appearred in the internet, and these videos increase in the mode of geometric series, become the main flow of the issue of information in the internet and amusement.In the face of the multi-medium data of magnanimity, how finding out similar even identical video from the magnanimity video becomes a hot research problem fast.
There are a lot of weak points in traditional text based retrieve video method.It mainly finishes the retrieval of video being carried out similar video by video periphery text is carried out index, but video content is different from content of text, have very big difference between video data and its semantic information, the mode that relies on peripheral text merely can not also have been ignored multimedia feature and information such as video visually-perceptible by the accurate description video content.On the other hand, adopt simple text based mode to carry out manual mark to massive video data in the network, workload is huge, marks video simultaneously and has certain subjectivity, for same section video different people different understanding may be arranged.Cause the accuracy rate of similarity retrieval very low thus, effect is difficult to be further improved.
Summary of the invention
The purpose of this invention is to provide a kind of method and system of similar video retrieval, improved similar video effectiveness of retrieval and accuracy.
A kind of method of similar video retrieval, this method comprises:
Utilize the color space model to calculate the eigenvector of each key frame of video set in the video library;
Shine upon described eigenvector by hash function, according to index building between the vector value location after the mapping;
According to the numbering of determining manipulative indexing between each vector value location after each motion vector mapping of video to be retrieved, extract the video information of all eigenvector correspondences under the manipulative indexing, carry out calculation of similarity degree according to the number of similar key frame in the video that extracts under video to be retrieved and the manipulative indexing;
To export as result for retrieval greater than the video of the result of calculation correspondence of threshold value.
A kind of system of similar video retrieval, this system comprises:
The eigenvector computing module is used for utilizing the color space model to calculate the eigenvector of each key frame of video set of video library;
The index construct module is used for shining upon described eigenvector by hash function, according to index building between the vector value location after the mapping;
Similarity calculation module, be used for according to the numbering of determining manipulative indexing between each vector value location after each motion vector mapping of video to be retrieved, extract the video information of all eigenvector correspondences under the manipulative indexing, carry out calculation of similarity degree according to the number of similar key frame in the video that extracts under video to be retrieved and the manipulative indexing;
The result for retrieval output module is used for and will exports as result for retrieval greater than the video of the result of calculation correspondence of threshold value.
As seen from the above technical solution provided by the invention, by the contents extraction eigenvector of video, utilize these eigenvectors to set up index, and carry out calculation of similarity degree with video to be retrieved based on this; Can efficiently reach the retrieval that in the magnanimity video library, realizes similar video accurately.
Description of drawings
In order to be illustrated more clearly in the technical scheme of the embodiment of the invention, the accompanying drawing of required use is done to introduce simply in will describing embodiment below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite of not paying creative work, can also obtain other accompanying drawings according to these accompanying drawings.
The process flow diagram of the method for a kind of similar video retrieval that Fig. 1 provides for the embodiment of the invention one;
The process flow diagram of a kind of clustering algorithm that Fig. 2 provides for the embodiment of the invention one;
The synoptic diagram of the system of a kind of similar video retrieval that Fig. 3 provides for the embodiment of the invention two;
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on embodiments of the invention, those of ordinary skills belong to protection scope of the present invention not making the every other embodiment that obtains under the creative work prerequisite.
Embodiment one
The process flow diagram of the method for a kind of similar video retrieval that Fig. 1 provides for the embodiment of the invention one.As shown in Figure 1, mainly comprise the steps:
Because the otherness of consecutive frame is little in the video, if adopt constant duration to extract the method for key frame, extracts the too high meeting of frequency and cause characteristic amount tremendous influence feature extraction speed; And the difference of consecutive frame is little, and similarity is too high, causes the feature redundancy; But, if extract underfrequency, though thereby can reduce the characteristic amount, accelerate the accuracy rate that general characteristic that characteristic set that feature extraction speed extracts can not react video has reduced retrieval comprehensively.
Therefore, all extract key frame earlier for each video data in the video library, for example, adopt clustering algorithm, it is by judging whether the feature between the adjacent image frame acute variation takes place, finish the rim detection task of video lens, namely with the comprehensive reflecting video general characteristic of characteristic set of simplifying most.
As shown in Figure 2, clustering algorithm mainly comprises the steps:
This step based on frame pitch from algorithm, judge whether content exists sudden change between the consecutive frame, consider that there is noise information in some frames of video beginning, so with its filtration, extract frame of video since the n frame, and centered by this frame the distance of each frame of calculated for subsequent and n frame.
If the frame pitch of n+i frame and n frame is from greater than threshold value, then the n+i frame is the frame that suddenlys change with respect to the n frame, as same classification, and selects the n frame as key-frame extraction the former frame (n+i-1) of n frame to the n+i.
Whether step 114, judgement residue frame number be less than S.
If current video residue frame number less than S(for example is, 2), then finish the extraction operation of key frame, otherwise, can be according to repeating above-mentioned steps until remaining frame number less than S.
Choose the key frame that obtains as can be seen the cluster from above-mentioned steps and can reflect preferably that the content of this section video lens and computation complexity are low.Need to prove that only for giving an example, the user also can use additive method to select representative key frame from video to the clustering algorithm of introducing above in real work.
Exemplary, can pass through the brightness of HSV(tone saturation degree) color space model calculated characteristics vector, the ability with abundant reflection color space distribution and information change situation strengthens the discrimination between the picture, improves retrieval performance.
After passing through the image transitions of hsv color spatial model with key frame, its each color of pixel tone h, saturation degree s, brightness v value representation.
Be to improve the accuracy of color model, can be with color close with black, white in the image respectively as treating with a kind of color, i.e. (1) black region: the color of all v＜15% all is included into black, makes h=0, s=0, v=0; (2) white portion: all s＜10% and v〉80% color is included into white, makes h=0, s=0, v=1; (3) colored region: be positioned at black region and white portion color in addition, its h, s, the v value remains unchanged.
Further, in order to reduce the dimension of histogram vectors, compute histograms is again carried out after the suitable quantification in the HSV space, reduce calculated amount.For example, can be with h, s, these 3 components of v carry out the quantification of unequal interval according to people's color-aware, produce the quantitative criteria table, extract the h of each pixel color in the image more respectively, s, the v value, obtain color h after each pixel quantizes, s, v value according to quantization table, count whole picture hsv color histogram vectors, i.e. the eigenvector of this key frame.
Video file in the database is more, even filtered out more redundant data by top two steps, but calculated amount is still bigger.Therefore, need index building, further reduce calculated amount.
Exemplary, can adopt the distributed LSH(local sensitivity Hash based on the internal memory constraint) structure carries out the structure of index.
Distributed LSH structure adopts the Master-Slaver(MS master-slave) framework.Master is responsible for safeguarding the partition strategy of LSH table, and Slaver(can be used as memory node) be responsible for safeguarding the data in the LSH table.
By the Hash mapping function, eigenvector is mapped as a n dimensional vector n, be stored in the Hash table of corresponding slaver maintenance.Be example with eigenvector b, master at first divides the Hash table numerical range that each slaver safeguards, with the conversion of eigenvector b process hash function, be mapped to its value of the one-dimensional space and be C, according to interval under the Hash table data area of each slaver node maintenance retrieval C, in the Hash bucket under being assigned to by the value of master after with this vector and mapping under the slaver.Each Hash bucket all has independently numbering, and all comprises the numbering of video and the frame of each eigenvector correspondence in each Hash bucket.
After each eigenvector is all by aforesaid operations, then finish the structure of distributed LSH configuration index.
In addition, can safeguard the data message in the Hash bucket and be kept in the internal memory by many data servers, overcome dimension disaster preferably, reduce the complicacy of retrieval, fast and reliable.
Retrieving mainly comprises the steps: 1) video to be retrieved is handled acquisition characteristic of correspondence vector according to the mode of step 11-step 12.2) with each eigenvector of video to be retrieved through hash function mapping, locate slaver and Hash bucket numbering under it according to the value after the mapping by master.3) extract to determine the video information of all eigenvector correspondences in the Hash bucket of numbering, and add up the number of the key frame that each video comprises respectively.4) calculate the similarity of video under the eigenvector that extracts in video to be retrieved and the Hash bucket according to the number of similar key frame in the video that extracts under video to be retrieved and the manipulative indexing and in conjunction with following formula:
Wherein, sim (V
i, V
j) ∈ [0,1]; | KF
i| be video V to be retrieved
iQuantity of key frames; | KF
j| be video V under the eigenvector that extracts in the Hash bucket
jQuantity of key frames; | KF
i∩ KF
j| expression V
iAnd V
jThe number of similar key frame.
Below as can be known, based on distributed LSH index structure non-data object similar, that can not become the result is filtered out, can further reduce calculated amount.
Can calculate the degree of correlation of each video in video to be retrieved and the current Hash bucket according to above-mentioned formula, the result is more big, and the degree of correlation that shows is more high, therefore, will export as result for retrieval greater than the video of the result of calculation correspondence of threshold value.
The embodiment of the invention is by using the Visual Feature Retrieval Process technology, and based on the distributed hash structure of internal memory constraint the high dimensional feature vector is set up index, according to the similar key frame set of returning video to be retrieved is carried out similarity and calculate, realized the quick retrieval of magnanimity similar video.
Through the above description of the embodiments, those skilled in the art can be well understood to above-described embodiment and can realize by software, also can realize by the mode that software adds necessary general hardware platform.Based on such understanding, the technical scheme of above-described embodiment can embody with the form of software product, it (can be CD-ROM that this software product can be stored in a non-volatile memory medium, USB flash disk, portable hard drive etc.) in, comprise some instructions with so that computer equipment (can be personal computer, server, the perhaps network equipment etc.) carry out the described method of each embodiment of the present invention.
Embodiment two
The synoptic diagram of the system of a kind of similar video retrieval that Fig. 4 provides for the embodiment of the invention two as shown in Figure 4, mainly comprises:
Eigenvector computing module 41 is used for utilizing the color space model to calculate the eigenvector of each key frame of video set of video library;
Index construct module 42 is used for shining upon described eigenvector by hash function, according to index building between the vector value location after the mapping;
Similarity calculation module 43, according to the numbering of determining manipulative indexing between each vector value location after each motion vector mapping of video to be retrieved, extract the video information of all eigenvector correspondences under the manipulative indexing, carry out calculation of similarity degree according to the number of similar key frame in the video that extracts under video to be retrieved and the manipulative indexing;
Result for retrieval output module 44 is used for and will exports as result for retrieval greater than the video of the result of calculation correspondence of threshold value.
Described eigenvector computing module 41 comprises:
Key frame set extraction module 411, be used for extracting each the key frame of video set of described video library by clustering algorithm, and this module comprises: cluster module 412, be used for extracting frame of video since the n frame, and centered by this frame each frame of calculated for subsequent and its frame pitch from; When the frame pitch of n+i frame and n frame from greater than threshold value the time, then n frame to the n+i-1 frame constitutes a cluster, with the n frame as key frame.
Described eigenvector computing module 41 comprises:
Color of image modular converter 412 is used for utilizing tone saturation degree brightness hsv color model that the color of image of key frame is changed;
Characteristics of image vector computing module 413, the image after being used for utilizing the HSV space to color conversion carries out unequal interval and quantizes, and carries out the eigenvector that histogram calculation obtains image.
Described index construct module 42 comprises:
Vector value acquisition module 421 is used for eigenvector is mapped to the one-dimensional space by hash function, obtains corresponding vector value;
Eigenvector distribution module 422 is used for determining affiliated memory node numbering according to the interval at vector value place, and eigenvector is assigned in the Hash bucket of this memory node correspondence; The information that comprises video and the frame of each eigenvector correspondence in the described Hash bucket.
Described similarity calculation module 43 comprises:
Information extraction modules 431 is used for according to the memory node under determining between each vector value location after each motion vector mapping of video to be retrieved and the numbering of corresponding Hash bucket, and extracts the information of the corresponding video of all eigenvectors in the reference numeral Hash bucket;
Video similarity calculation module 432 be used for to be calculated the similarity of video under video to be retrieved and the eigenvector that extracts from the Hash bucket, and its formula is:
Wherein, | KF
i| be video V to be retrieved
iQuantity of key frames; | KF
j| be video V under the eigenvector that extracts in the Hash bucket
jQuantity of key frames; | KF
i∩ KF
j| expression V
iAnd V
jThe number of similar key frame.
Need to prove, have a detailed description among the specific implementation of the function that each processing unit that comprises in the said apparatus is realized each embodiment in front, so here repeat no more.
The embodiment of the invention is by using the Visual Feature Retrieval Process technology, and based on the distributed hash structure of internal memory constraint to the classification of high dimensional feature vector and set up index, according to the similar key frame set of returning video to be retrieved is carried out similarity and calculate, realized the quick retrieval of magnanimity similar video.
The those skilled in the art can be well understood to, be the convenience described and succinct, only the division with above-mentioned each functional module is illustrated, in the practical application, can as required the above-mentioned functions distribution be finished by different functional modules, the inner structure that is about to device is divided into different functional modules, to finish all or part of function described above.
The above; only for the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, and anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.
Claims (10)
1. the method for similar video retrieval is characterized in that this method comprises:
Utilize the color space model to calculate the eigenvector of each key frame of video set in the video library;
Shine upon described eigenvector by hash function, according to index building between the vector value location after the mapping;
According to the numbering of determining manipulative indexing between each vector value location after each motion vector mapping of video to be retrieved, extract the video information of all eigenvector correspondences under the manipulative indexing, carry out calculation of similarity degree according to the number of similar key frame in the video that extracts under video to be retrieved and the manipulative indexing;
To export as result for retrieval greater than the video of the result of calculation correspondence of threshold value.
2. method according to claim 1 is characterized in that, this method also comprises the step of extracting the key frame of video set by clustering algorithm, and its step comprises:
Extract frame of video since the n frame, and centered by this frame each frame of calculated for subsequent and its frame pitch from;
When the frame pitch of n+i frame and n frame from greater than threshold value the time, then n frame to the n+i-1 frame constitutes a cluster, and with the n frame as key frame.
3. method according to claim 1 is characterized in that, the described step of utilizing the color space model to calculate the eigenvector of each key frame of video set in the video library comprises:
Utilize tone saturation degree brightness hsv color model that the color of image in the key frame is changed;
Image after utilizing the HSV space to color conversion carries out unequal interval and quantizes, and carries out the eigenvector that histogram calculation obtains image.
4. method according to claim 1 is characterized in that, described step according to index building between the vector value location after the mapping comprises:
Eigenvector is mapped to the one-dimensional space by hash function, obtains corresponding vector value;
Determine affiliated memory node numbering according to the interval at vector value place, and eigenvector is assigned in the Hash bucket of this memory node correspondence; The information that comprises video and the frame of each eigenvector correspondence in the described Hash bucket.
5. method according to claim 4 is characterized in that, described step of carrying out similarity calculating comprises:
According to the memory node under determining between each vector value location after the mapping of each motion vector of video to be retrieved and the numbering of corresponding Hash bucket, and extract the information of the corresponding videos of all eigenvectors in the reference numeral Hash bucket;
The similarity of video under the eigenvector that calculates video to be retrieved and from the Hash bucket, extract, its formula is:
Wherein, | KF
i| be video V to be retrieved
iQuantity of key frames; | KF
j| be video V under the eigenvector that extracts in the Hash bucket
jQuantity of key frames; | KF
i∩ KF
j| expression V
iAnd V
jThe number of similar key frame.
6. the system of similar video retrieval is characterized in that this system comprises:
The eigenvector computing module is used for utilizing the color space model to calculate the eigenvector of each key frame of video set of video library;
The index construct module is used for shining upon described eigenvector by hash function, according to index building between the vector value location after the mapping;
Similarity calculation module, be used for according to the numbering of determining manipulative indexing between each vector value location after each motion vector mapping of video to be retrieved, extract the video information of all eigenvector correspondences under the manipulative indexing, carry out calculation of similarity degree according to the number of similar key frame in the video that extracts under video to be retrieved and the manipulative indexing;
The result for retrieval output module is used for and will exports as result for retrieval greater than the video of the result of calculation correspondence of threshold value.
7. system according to claim 6 is characterized in that, described eigenvector computing module comprises:
Key frame set extraction module be used for extracting the key frame of video set by clustering algorithm, and this module comprises: the cluster module, be used for extracting frame of video since the n frame, and centered by this frame each frame of calculated for subsequent and its frame pitch from; When the frame pitch of n+i frame and n frame from greater than threshold value the time, then n frame to the n+i-1 frame constitutes a cluster, and with the n frame as key frame.
8. system according to claim 6 is characterized in that, described eigenvector computing module comprises:
The color of image modular converter is used for utilizing tone saturation degree brightness hsv color model that the color of image of key frame is changed;
Characteristics of image vector computing module, the image after being used for utilizing the HSV space to color conversion carries out unequal interval and quantizes, and carries out the eigenvector that histogram calculation obtains image.
9. system according to claim 6 is characterized in that, described index construct module comprises:
The vector value acquisition module is used for eigenvector is mapped to the one-dimensional space by hash function, obtains corresponding vector value;
The eigenvector distribution module is used for determining affiliated memory node numbering according to the interval at vector value place, and eigenvector is assigned in the Hash bucket of this memory node correspondence; The information that comprises video and the frame of each eigenvector correspondence in the described Hash bucket.
10. system according to claim 9 is characterized in that, described similarity calculation module comprises:
Information extraction modules is used for according to the memory node under determining between each vector value location after each motion vector mapping of video to be retrieved and the numbering of corresponding Hash bucket, and extracts the information of the corresponding video of all eigenvectors in the reference numeral Hash bucket;
The video similarity calculation module be used for to be calculated the similarity of video under video to be retrieved and the eigenvector that extracts from the Hash bucket, and its formula is:
Wherein, | KF
i| be video V to be retrieved
iQuantity of key frames; | KF
j| be video V under the eigenvector that extracts in the Hash bucket
jQuantity of key frames; | KF
i∩ KF
j| expression V
iAnd V
jThe number of similar key frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100340117A CN103257992A (en) | 2013-01-29 | 2013-01-29 | Method and system for retrieving similar videos |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100340117A CN103257992A (en) | 2013-01-29 | 2013-01-29 | Method and system for retrieving similar videos |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103257992A true CN103257992A (en) | 2013-08-21 |
Family
ID=48961917
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013100340117A Pending CN103257992A (en) | 2013-01-29 | 2013-01-29 | Method and system for retrieving similar videos |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103257992A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103984778A (en) * | 2014-06-06 | 2014-08-13 | 北京金山网络科技有限公司 | Video retrieval method and video retrieval system |
| GB2518876A (en) * | 2013-10-04 | 2015-04-08 | Nokia Corp | Method, apparatus and computer program product for similarity determination in multimedia content |
| CN104504101A (en) * | 2014-12-30 | 2015-04-08 | 北京奇艺世纪科技有限公司 | Determination method and device of similarity videos |
| WO2016000509A1 (en) * | 2014-06-30 | 2016-01-07 | 华为技术有限公司 | Data filtering method, and data filter construction method and apparatus |
| CN105681898A (en) * | 2015-12-31 | 2016-06-15 | 北京奇艺世纪科技有限公司 | Similar video and pirated video detection method and device |
| CN106156284A (en) * | 2016-06-24 | 2016-11-23 | 合肥工业大学 | Video retrieval method is closely repeated based on random the extensive of various visual angles Hash |
| CN106375847A (en) * | 2015-07-23 | 2017-02-01 | 无锡天脉聚源传媒科技有限公司 | Template generation method, template generation device, video updating method and video updating device |
| CN106372092A (en) * | 2015-07-23 | 2017-02-01 | 无锡天脉聚源传媒科技有限公司 | Method and device for generating template, and video update method and device |
| CN106557545A (en) * | 2016-10-19 | 2017-04-05 | 北京小度互娱科技有限公司 | Video retrieval method and device |
| CN106982389A (en) * | 2017-03-17 | 2017-07-25 | 腾讯科技（深圳）有限公司 | Video type recognition methods and device |
| CN110222594A (en) * | 2019-05-20 | 2019-09-10 | 厦门能见易判信息科技有限公司 | Pirate video recognition methods and system |
| CN110427895A (en) * | 2019-08-06 | 2019-11-08 | 李震 | A kind of video content similarity method of discrimination based on computer vision and system |
| CN110502664A (en) * | 2019-08-27 | 2019-11-26 | 腾讯科技（深圳）有限公司 | Video tab indexes base establishing method, video tab generation method and device |
| CN110955800A (en) * | 2018-09-26 | 2020-04-03 | 传线网络科技（上海）有限公司 | Video retrieval method and device |
| CN111064865A (en) * | 2019-12-31 | 2020-04-24 | 中国人民解放军国防科技大学 | Background activity noise filter of dynamic vision sensor and processor |
| CN111507260A (en) * | 2020-04-17 | 2020-08-07 | 重庆邮电大学 | Video similarity rapid detection method and detection device |
| CN113254701A (en) * | 2021-06-21 | 2021-08-13 | 北京电信易通信息技术股份有限公司 | Video data index construction method and system based on FOV model |
| CN113536031A (en) * | 2021-06-17 | 2021-10-22 | 北京百度网讯科技有限公司 | Video searching method and device, electronic equipment and storage medium |
| CN114041165A (en) * | 2019-07-18 | 2022-02-11 | 华为技术有限公司 | Video similarity detection method, device and equipment |
| CN114676278A (en) * | 2022-03-22 | 2022-06-28 | 鲸灵云(杭州)智能有限公司 | User picture identification processing method and system based on private domain e-commerce platform and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5805746A (en) * | 1993-10-20 | 1998-09-08 | Hitachi, Ltd. | Video retrieval method and apparatus |
| JP2006060796A (en) * | 2004-07-21 | 2006-03-02 | Nippon Telegr & Teleph Corp <Ntt> | Video search apparatus, method, and program, and recording medium with the program stored thereon |
-
2013
- 2013-01-29 CN CN2013100340117A patent/CN103257992A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5805746A (en) * | 1993-10-20 | 1998-09-08 | Hitachi, Ltd. | Video retrieval method and apparatus |
| JP2006060796A (en) * | 2004-07-21 | 2006-03-02 | Nippon Telegr & Teleph Corp <Ntt> | Video search apparatus, method, and program, and recording medium with the program stored thereon |
Non-Patent Citations (4)
| Title |
|---|
| 姜兰池，沈国强，张国煊: "基于HSV分块颜色直方图的图像检索方法", 《机电工程》 * |
| 曹海傧，朱明，冯伟国: "一种基于分布式LSH的海量视频快速检索方法", 《中国科学院研究生院学报》 * |
| 朱映映，周洞汝: "一种基于视频聚类的关键帧提取方法", 《计算机工程》 * |
| 潘晓英: "一种基于核聚类的关键帧提取方法", 《微机发展》 * |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2518876A (en) * | 2013-10-04 | 2015-04-08 | Nokia Corp | Method, apparatus and computer program product for similarity determination in multimedia content |
| CN103984778A (en) * | 2014-06-06 | 2014-08-13 | 北京金山网络科技有限公司 | Video retrieval method and video retrieval system |
| US9755616B2 (en) | 2014-06-30 | 2017-09-05 | Huawei Technologies Co., Ltd. | Method and apparatus for data filtering, and method and apparatus for constructing data filter |
| WO2016000509A1 (en) * | 2014-06-30 | 2016-01-07 | 华为技术有限公司 | Data filtering method, and data filter construction method and apparatus |
| CN104504101A (en) * | 2014-12-30 | 2015-04-08 | 北京奇艺世纪科技有限公司 | Determination method and device of similarity videos |
| CN104504101B (en) * | 2014-12-30 | 2018-10-30 | 北京奇艺世纪科技有限公司 | A kind of determination method and device of similar video |
| CN106375847A (en) * | 2015-07-23 | 2017-02-01 | 无锡天脉聚源传媒科技有限公司 | Template generation method, template generation device, video updating method and video updating device |
| CN106372092A (en) * | 2015-07-23 | 2017-02-01 | 无锡天脉聚源传媒科技有限公司 | Method and device for generating template, and video update method and device |
| CN105681898B (en) * | 2015-12-31 | 2018-10-30 | 北京奇艺世纪科技有限公司 | A kind of detection method and device of similar video and pirate video |
| CN105681898A (en) * | 2015-12-31 | 2016-06-15 | 北京奇艺世纪科技有限公司 | Similar video and pirated video detection method and device |
| CN106156284A (en) * | 2016-06-24 | 2016-11-23 | 合肥工业大学 | Video retrieval method is closely repeated based on random the extensive of various visual angles Hash |
| CN106557545A (en) * | 2016-10-19 | 2017-04-05 | 北京小度互娱科技有限公司 | Video retrieval method and device |
| CN106557545B (en) * | 2016-10-19 | 2020-08-07 | 北京小度互娱科技有限公司 | Video retrieval method and device |
| CN106982389A (en) * | 2017-03-17 | 2017-07-25 | 腾讯科技（深圳）有限公司 | Video type recognition methods and device |
| CN110955800A (en) * | 2018-09-26 | 2020-04-03 | 传线网络科技（上海）有限公司 | Video retrieval method and device |
| CN110222594B (en) * | 2019-05-20 | 2021-11-16 | 厦门能见易判信息科技有限公司 | Pirated video identification method and system |
| CN110222594A (en) * | 2019-05-20 | 2019-09-10 | 厦门能见易判信息科技有限公司 | Pirate video recognition methods and system |
| CN114041165A (en) * | 2019-07-18 | 2022-02-11 | 华为技术有限公司 | Video similarity detection method, device and equipment |
| CN110427895A (en) * | 2019-08-06 | 2019-11-08 | 李震 | A kind of video content similarity method of discrimination based on computer vision and system |
| CN110502664A (en) * | 2019-08-27 | 2019-11-26 | 腾讯科技（深圳）有限公司 | Video tab indexes base establishing method, video tab generation method and device |
| CN111064865A (en) * | 2019-12-31 | 2020-04-24 | 中国人民解放军国防科技大学 | Background activity noise filter of dynamic vision sensor and processor |
| CN111064865B (en) * | 2019-12-31 | 2021-08-31 | 中国人民解放军国防科技大学 | Background activity noise filter of dynamic vision sensor and processor |
| CN111507260A (en) * | 2020-04-17 | 2020-08-07 | 重庆邮电大学 | Video similarity rapid detection method and detection device |
| CN111507260B (en) * | 2020-04-17 | 2022-08-05 | 重庆邮电大学 | Video similarity rapid detection method and detection device |
| CN113536031A (en) * | 2021-06-17 | 2021-10-22 | 北京百度网讯科技有限公司 | Video searching method and device, electronic equipment and storage medium |
| CN113254701A (en) * | 2021-06-21 | 2021-08-13 | 北京电信易通信息技术股份有限公司 | Video data index construction method and system based on FOV model |
| CN114676278A (en) * | 2022-03-22 | 2022-06-28 | 鲸灵云(杭州)智能有限公司 | User picture identification processing method and system based on private domain e-commerce platform and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103257992A (en) | Method and system for retrieving similar videos | |
| Kang | Affective content detection using HMMs | |
| CN104376003B (en) | A kind of video retrieval method and device | |
| CN107169106B (en) | Video retrieval method, device, storage medium and processor | |
| CN103347167A (en) | Surveillance video content description method based on fragments | |
| CN104520875A (en) | A method and an apparatus for the extraction of descriptors from video content, preferably for search and retrieval purpose | |
| CN104679818A (en) | Video keyframe extracting method and video keyframe extracting system | |
| CN103631932A (en) | Method for detecting repeated video | |
| CN105678244B (en) | A kind of near video search method based on improved edit-distance | |
| Arifin et al. | A computation method for video segmentation utilizing the pleasure-arousal-dominance emotional information | |
| Noman et al. | Object detection techniques: Overview and performance comparison | |
| CN112532643A (en) | Deep learning-based traffic anomaly detection method, system, terminal and medium | |
| CN104410867A (en) | Improved video shot detection method | |
| Fernandes et al. | Low power affordable and efficient face detection in the presence of various noises and blurring effects on a single-board computer | |
| CN108121806A (en) | One kind is based on the matched image search method of local feature and system | |
| CN107203638A (en) | Monitor video processing method, apparatus and system | |
| CN105404682B (en) | A kind of book retrieval method based on digital image content | |
| CN110490250A (en) | A kind of acquisition methods and device of artificial intelligence training set | |
| CN117119253B (en) | High-quality video frame extraction method for target object | |
| CN110287369A (en) | A kind of semantic-based video retrieval method and system | |
| Li et al. | PFYOLOv4: An improved small object pedestrian detection algorithm | |
| Yang et al. | CEFusion: An infrared and visible image fusion network based on cross-modal multi-granularity information interaction and edge guidance | |
| Aziz et al. | Multi-level refinement feature pyramid network for scale imbalance object detection | |
| Zhang et al. | An improved PAM clustering algorithm based on initial clustering centers | |
| Hu et al. | STRNN: End-to-end deep learning framework for video partial copy detection |
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: 20130821 |
|
| RJ01 | Rejection of invention patent application after publication |