CN113542348B

CN113542348B - Image data transmission method and device

Info

Publication number: CN113542348B
Application number: CN202110590068.XA
Authority: CN
Inventors: 张亮; 陈超允; 李林
Original assignee: Wuhan Kuangshi Jinzhi Technology Co ltd; Beijing Megvii Technology Co Ltd
Current assignee: Wuhan Kuangshi Jinzhi Technology Co ltd; Beijing Megvii Technology Co Ltd
Priority date: 2021-05-27
Filing date: 2021-05-27
Publication date: 2022-09-06
Anticipated expiration: 2041-05-27
Also published as: CN113542348A

Abstract

The invention provides an image data transmission method and device, which relate to the technical field of computers and comprise the steps of determining attribute information of at least one algorithm adopted by first operation data corresponding to a first image in a first view library to be subscribed, wherein the first view library comprises the first image and the first operation data; determining whether first matching operation data matched with an algorithm provided by a manufacturer exists in the first operation data based on attribute information of at least one algorithm adopted by the first operation data; and if the first operation data contains first matching operation data matched with the algorithm provided by the manufacturer, sending a first subscription request to a first server where the first view library is located, wherein the first subscription request is used for subscribing the first image and the first matching operation data. When the first view library to be subscribed has locally required operation data, the first view library to be subscribed can be subscribed together, so that the resource consumption of local transportation can be reduced, and the resource utilization rate is improved.

Description

Image data transmission method and device

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a method and an apparatus for transmitting image data.

Background

In general, image features need to be extracted from images, and similarity between images needs to be measured by distances between image features, so as to implement image applications such as image retrieval. Image features extracted by different algorithms cannot be compared, for example, the image 1 extracts the image feature 1 by using the algorithm a, the image 2 extracts the image feature 2 by using the algorithm B, the similarity between the image 1 and the image 2 can be measured by using the distance between the image feature 1 and the image feature 2 only when the algorithm a and the algorithm B are compatible, and the distance between the image feature 1 and the image feature 2 cannot be used for measuring the similarity between the image 1 and the image 2 when the algorithm a and the algorithm B are incompatible.

At present, image data is applied more and more across levels and networks, for example, an upper level needs to collect images acquired by a lower level, the lower level needs to retrieve images transmitted by the upper level, and then, for example, image synchronization is performed between different networks, so that image data needs to be transmitted between the upper level and the lower level (for example, between a subsidiary company and a main company) and between different networks. However, algorithms used by the upper level and the lower level are often overlapped but not completely the same, if data transmission is performed, only the image itself is transmitted, and the image features are extracted again by the receiving side, and there may be a case that the image is extracted once by using a certain algorithm at the transmitting side and is extracted once again by using the same algorithm at the receiving side, so that certain calculation power is wasted. If the image + characteristics are transmitted uniformly during data transmission, the characteristics transmitted to the receiver by the sender may not be used, and the receiver needs to extract the characteristics again according to the image, so that bandwidth is occupied.

Disclosure of Invention

The invention aims to provide an image data transmission method and device, which are used for relieving the technical problem of resource waste in the prior art.

In a first aspect, the present invention provides an image data transmission method, including:

determining attribute information of at least one algorithm adopted by first operation data corresponding to a first image in a first view library to be subscribed, wherein the first view library comprises the first image and the first operation data;

determining whether first matching operation data matched with an algorithm provided by a manufacturer exists in the first operation data based on attribute information of at least one algorithm adopted by the first operation data; if first matching operation data matched with an algorithm provided by a manufacturer exist in the first operation data, sending a first subscription request to a first server where the first view library is located, wherein the first subscription request is used for subscribing the first image and the first matching operation data.

In an optional implementation, the method further comprises:

if first matching operation data matched with an algorithm provided by a manufacturer does not exist in the first operation data, sending a second subscription request to the first server, wherein the second subscription request is used for subscribing the first image;

receiving gallery update data sent by the first server in response to the second subscription request, the gallery update data including an updated first image;

and calculating the updated first image based on the algorithm provided by the manufacturer to obtain first target calculation data corresponding to the updated first image.

In an optional implementation, before the determining attribute information of at least one algorithm used for first operation data corresponding to a first image in a first view library to be subscribed, the method further includes:

determining whether a first image in a first view library to be subscribed corresponds to first operation data or not;

if the first image does not correspond to the first operation data, sending a second subscription request to the first server, wherein the second subscription request is used for subscribing the first image;

In an optional implementation, the method further comprises:

sending the first target operational data to the first server, so that the first server updates first operational data corresponding to the updated first image in the first view library based on the first target operational data.

In an alternative implementation, the attribute information of the algorithm includes an algorithm manufacturer identifier and an algorithm version identifier; the operational data includes image features and/or image attributes.

In a second aspect, an image data transmission method is provided, including:

determining attribute information of at least one algorithm adopted by second operation data corresponding to a second image in a second view library to be subscribed, wherein the second view library comprises the second image and the second operation data;

determining whether second matching operation data matched with a view library algorithm exists in the second operation data based on attribute information of at least one algorithm adopted by the second operation data;

if second matching operation data matched with the view library algorithm exists in the second operation data, sending a third subscription request to a second server where the second view library is located, wherein the third subscription request is used for subscribing the second image and the second matching operation data; receiving gallery update data sent by the second server in response to the third subscription request, wherein the gallery update data comprises an updated second image, second matching operation data corresponding to the updated second image and attribute information of an algorithm adopted by the second matching operation data;

and updating the current-level view library based on the gallery update data.

In an optional implementation, the method is applied to a third server, where the lower level servers corresponding to the third server include a plurality of lower level servers, each of the lower level servers corresponds to one lower level view library, the second server is one or more of the lower level servers, and the second view library is one or more of the lower level view libraries.

In an optional implementation, the method further comprises:

if second matching operation data matched with the view library algorithm does not exist in the second operation data, sending a fourth subscription request to a second server where the second view library is located, wherein the fourth subscription request is used for subscribing the second image; and receiving gallery update data sent by the second server in response to the fourth subscription request, the gallery update data including an updated second image;

sending the updated second image to a server provided by an algorithm manufacturer so that the server provided by the algorithm manufacturer can calculate the updated second image based on a manufacturer algorithm matched with a view base algorithm to obtain second target calculation data corresponding to the updated second image;

receiving the second target operation data from a server provided by the algorithm manufacturer and attribute information of the manufacturer algorithm;

and updating second operation data corresponding to the updated second image in the second view library based on the second target operation data.

In a third aspect, there is provided an image data transmission apparatus comprising:

the device comprises a determining module, a processing module and a processing module, wherein the determining module is used for determining attribute information of at least one algorithm adopted by first operation data corresponding to a first image in a first view library to be subscribed, and the first view library comprises the first image and the first operation data;

the matching module is used for determining whether first matching operation data matched with an algorithm provided by a manufacturer exists in the first operation data based on attribute information of at least one algorithm adopted by the first operation data;

and the subscription module is used for sending a first subscription request to a first server where the first view library is located if first matching operation data matched with an algorithm provided by a manufacturer exists in the first operation data, wherein the first subscription request is used for subscribing the first image and the first matching operation data.

In an optional implementation, the method further comprises:

the subscription module is further configured to send a second subscription request to the first server if first matching operation data matching an algorithm provided by a vendor does not exist in the first operation data, where the second subscription request is used to subscribe to the first image;

a receiving module, configured to receive gallery update data sent by the first server in response to the second subscription request, where the gallery update data includes an updated first image;

and the operation module is used for operating the updated first image based on the manufacturer algorithm to obtain first target operation data corresponding to the updated first image.

In an optional implementation, the method further comprises:

the determining module is further used for determining whether a first image in a first view library to be subscribed corresponds to first operation data;

the subscription module is further configured to send a second subscription request to the first server if the first image does not correspond to the first operation data, where the second subscription request is used to subscribe to the first image;

and calculating the updated first image based on the manufacturer algorithm to obtain first target calculation data corresponding to the updated first image.

In an optional implementation, the method further comprises:

a sending module, configured to send the first target operation data to the first server, so that the first server updates, based on the first target operation data, first operation data corresponding to the updated first image in the first view library.

In a fourth aspect, there is provided an image data transmission apparatus comprising:

the determining module is used for determining attribute information of at least one algorithm adopted by second operation data corresponding to a second image in a second view library to be subscribed, wherein the second view library comprises the second image and the second operation data;

the matching module is used for determining whether second matching operation data matched with the view base algorithm exists in the second operation data or not based on the attribute information of at least one algorithm adopted by the second operation data;

a subscription module, configured to send a third subscription request to a second server in which the second view library is located if second matching operation data matching the view library algorithm exists in the second operation data, where the third subscription request is used to subscribe to the second image and the second matching operation data; receiving gallery update data sent by the second server in response to the third subscription request, wherein the gallery update data comprises an updated second image, second matching operation data corresponding to the updated second image and attribute information of an algorithm adopted by the second matching operation data;

and the updating module is used for updating the current-level view library based on the gallery updating data.

In an optional implementation, the apparatus is applied to a third server, where the third server corresponds to a plurality of lower level servers, each of the lower level servers corresponds to a lower level view library, the second server is one or more of the lower level servers, and the second view library is one or more of the lower level view libraries.

In an optional implementation, the method further comprises:

the subscription module is further configured to send a fourth subscription request to a second server where the second view library is located if second matching operation data matching the view library algorithm does not exist in the second operation data, where the fourth subscription request is used to subscribe to the second image; and receiving gallery update data sent by the second server in response to the fourth subscription request, the gallery update data including an updated second image;

the sending module is used for sending the updated second image to a server provided by an algorithm manufacturer so that the server provided by the algorithm manufacturer can calculate the updated second image based on a manufacturer algorithm matched with the view base algorithm to obtain second target calculation data corresponding to the updated second image;

a receiving module, configured to receive the second target operation data and attribute information of the vendor algorithm from a server provided by the algorithm vendor;

the updating module is further configured to update second operation data corresponding to the updated second image in the second view library based on the second target operation data.

In a fifth aspect, the present invention provides an electronic device, comprising: a processing device and a storage device;

the storage means has stored thereon a computer program which, when executed by the processor, performs the method according to any of the preceding embodiments for image data transmission.

In a sixth aspect, the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of any one of the preceding embodiments.

The invention provides an image data transmission method and device. Determining attribute information of at least one algorithm adopted by first operation data corresponding to a first image in a first view library to be subscribed, wherein the first view library comprises the first image and the first operation data; determining whether first matching operation data matched with an algorithm provided by a manufacturer exists in the first operation data based on attribute information of at least one algorithm adopted by the first operation data; if first matching operation data matched with an algorithm provided by a manufacturer exist in the first operation data, sending a first subscription request to a first server where the first view library is located, wherein the first subscription request is used for subscribing the first image and the first matching operation data. Therefore, when the operation data matched with the algorithm provided by the manufacturer exists in the first view library to be subscribed, the subscription can be carried out together, the waste of the calculation power caused by repeated calculation of the operation data is avoided, meanwhile, the transmission of unmatched operation data can be avoided, the network bandwidth consumption is reduced, and the resource utilization rate is improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.

FIG. 1 is an exemplary electronic device for implementing the level view library method of an embodiment of the present invention;

FIG. 2 is a flow diagram of a method of a level view library according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a method of a level view library according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of another level view library method according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of another level view library method according to an embodiment of the present invention;

FIG. 6 is a flow diagram of another level view library method according to an embodiment of the present invention;

FIG. 7 is a schematic diagram of a level view library arrangement according to an embodiment of the present invention;

FIG. 8 is a schematic diagram of a level view library apparatus according to an embodiment of the present invention;

fig. 9 is a schematic structural diagram of an image data transmission apparatus according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

First, an exemplary electronic device 100 for implementing the image data transmission method of the embodiment of the present invention is described with reference to fig. 1.

As shown in FIG. 1, electronic device 100 includes one or more processing devices 102, one or more memory devices 104, an input device 106, and an output device 108, which are interconnected via a bus system 112 and/or other form of connection mechanism (not shown). It should be noted that the components and structure of the electronic device 100 shown in fig. 1 are exemplary only, and not limiting, and that the electronic device may have other components and structures as desired.

The processing device 102 may be a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 100 to perform desired functions.

Storage 104 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. Volatile memory can include, for example, Random Access Memory (RAM), cache memory (cache), and/or the like. The non-volatile memory may include, for example, Read Only Memory (ROM), a hard disk, flash memory, and the like. One or more computer program instructions may be stored on a computer-readable storage medium and executed by processing device 102 to implement the client functionality (implemented by a processor) of the embodiments of the invention described below and/or other desired functionality. Various applications and various data, such as various data used and/or generated by the applications, may also be stored in the computer-readable storage medium.

The input device 106 may be a device used by a user to input instructions and may include one or more of a keyboard, a mouse, a microphone, a touch screen, and the like.

The output device 108 may output various information (e.g., images or sounds) to the outside (e.g., a user), and may include one or more of a display, a speaker, and the like.

Exemplarily, an exemplary electronic device for implementing the image data transmission method according to the embodiment of the present invention may be implemented on a mobile terminal such as a server, a smartphone, a tablet computer, and the like.

In the embodiment of the invention, when the view library to be subscribed has the locally required operation data, the view library can be subscribed together, so that the resource consumption of local transportation can be reduced, and the resource utilization rate is improved.

In accordance with an embodiment of the present invention, there is provided an embodiment of an image data transmission method, it should be noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer executable instructions, and that while a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than that herein.

Fig. 2 is a flowchart of an image data transmission method according to an embodiment of the present invention, as shown in fig. 2, the method including the steps of:

s210, determining attribute information of at least one algorithm adopted by first operation data corresponding to a first image in a first view library to be subscribed, wherein the first view library comprises the first image and the first operation data.

The first view library may include a first image, first operation data corresponding to the first image, and attribute information of at least one algorithm employed by the first operation data.

The first operation data corresponding to the first image may include at least one type, and each type of the first operation data is obtained by an algorithm operation.

The attribute information of the algorithm comprises an algorithm manufacturer identifier and/or an algorithm version identifier; the operational data may include image features and/or image attributes.

The image feature refers to a feature obtained by performing feature extraction on the first image according to an algorithm. The image attribute may refer to a result of attribute recognition of the first image based on the extracted image feature. For example, the gender recognition of the human face in the first image can be performed based on the extracted image features, resulting in a gender recognition result, which can be considered as an image attribute.

The method can be applied to a server (also called algorithm server) provided by an algorithm manufacturer, and the algorithm manufacturer can provide algorithm service for the view library server. Typically, for a view library server, there will be at least one algorithm vendor that provides algorithm services to it. The algorithm service provided by the algorithm vendor for the view library may be an image retrieval service. The image retrieval process is generally a process of retrieving, from an image to be retrieved (Gallery image), the first N images having a similarity higher than a threshold value with respect to a Query image (Query image). To calculate the similarity between the two images, the same feature extraction algorithm or a compatible feature extraction algorithm is required to extract the features of the two images in advance. The algorithm service provided by the algorithm manufacturer for the view library may be that the view library has an image update (the updated image may be from a camera connected to the current-level view library or from an image transmitted from the lower-level view library), and the image update may also have features corresponding to the image (at this time, the image transmitted from the lower-level view library may have features corresponding to the image simultaneously). And the algorithm manufacturer server subscribes to the view library, responds to the update of the images and/or the image characteristics on the view library, and only acquires the updated images and extracts the characteristics of the updated images or acquires the updated images and the image characteristics corresponding to the updated images. When the algorithm manufacturer server obtains the updated image and performs feature extraction on the updated image, the image features obtained by feature extraction are also generally sent to the current-level view library. Therefore, the image and the image features are stored in the algorithm server, when the algorithm server acquires the query image and extracts the features of the query image, the image features of the query image can be compared with the image features of the image stored in the algorithm server to obtain the similarity of the image features and the image features, and then the retrieval result is obtained.

In the embodiment of the present invention, the view library server may include multiple levels, and the same algorithm manufacturer may provide algorithm services for different levels, or for different view library servers of the same level. Typically, the server provided by the algorithm vendor subscribes to only the view library in the level of view library server that it provides the service.

As shown in fig. 3, fig. 3 shows two levels of gallery, namely, an upper view gallery stored in the upper view gallery server and a lower view gallery including a lower a gallery (which may also be referred to as a lower a gallery), a lower B gallery (which may also be referred to as a lower B gallery), a lower C gallery (which may also be referred to as a lower C gallery), a lower D gallery (which may also be referred to as a lower D gallery) and a lower E gallery (which may also be referred to as a lower E gallery).

Also shown in FIG. 3 are 3 servers provided by algorithm vendors, namely, a server provided by algorithm vendor A, a server provided by algorithm vendor B, and a server provided by algorithm vendor C; the server provided by the algorithm manufacturer A can subscribe the lower-level A gallery, the lower-level D gallery and the lower-level E gallery; the server provided by the algorithm manufacturer B provides service for the lower-level B gallery, and the server provided by the algorithm manufacturer B can subscribe the lower-level B gallery; and the server provided by the algorithm manufacturer C provides service for the lower-level C gallery, and the server provided by the algorithm manufacturer C can subscribe the lower-level C gallery.

As one example, the multi-level view library server may include a lower level view library server and an upper level view library server. The lower level gallery can acquire the face picture and simultaneously acquire the characteristic value and the algorithm version of each algorithm (the peer algorithm server corresponding to the lower level view gallery extracts the characteristics of the picture to obtain the characteristic value and sends the characteristic value and the algorithm version to the lower level view gallery, so that the lower level view gallery acquires the characteristic value and the algorithm version while acquiring the picture). Therefore, when the higher-level view library summarizes resources, the pictures, the characteristics and the algorithm version of the lower-level view library can be summarized together. An algorithm server providing services for a superior view library server can subscribe the superior view library, specifically, can classify and subscribe in different regions, for example, if algorithm characteristics required by the algorithm server exist, the algorithm server can subscribe pictures and characteristics at the same time; if there are no algorithmic features required by the algorithm server, then only the picture may be subscribed to.

As another example, the multi-level view library server may belong to different networks, and peer cross-network transmission may subscribe to the eigenvalue and algorithm version of the algorithm; for example, when subscribing data to other networks, the first network may subscribe according to its own needs.

For the lower view library server, while uploading the face image to the upper view library server, the face operation data (e.g., face features and face attributes) corresponding to the face image and the attribute information (e.g., algorithm version) of the algorithm corresponding to the face operation data may be uploaded to the upper view library server together.

S220, determining whether first matching operation data matched with the algorithm provided by the manufacturer exists in the first operation data based on the attribute information of at least one algorithm adopted by the first operation data.

For example, three algorithms a, B and C are used for the first operation data corresponding to the first image in the current-level view library corresponding to the algorithm server, and the algorithm server runs the algorithm a, so that the part of the first operation data determined by using the algorithm a is usable by the algorithm server for image retrieval and comparison, and the part of the first operation data determined by using the algorithm B, C is not usable by the algorithm server for image retrieval and comparison. And then, the first operation data corresponding to the first image in the current-level view library corresponding to the algorithm server adopts two algorithms B and C, and the algorithm A runs on the algorithm server, so that the first operation data does not have a part which can be used for the algorithm server to carry out image retrieval comparison, and the algorithm server needs to calculate according to the first image by using the algorithm A to obtain the first operation data which can be used for the algorithm server to carry out image retrieval comparison.

If at least one first operation data corresponding to the first image exists, the current first operation data can be selected from the at least one first operation data corresponding to the first image in sequence, and the attribute information corresponding to the current first operation data is matched with the algorithm provided by the manufacturer until the matching is successful or the at least one first operation data corresponding to the first image is traversed.

And S230, if first matching operation data matched with the algorithm provided by the manufacturer exists in the first operation data, sending a first subscription request to a first server where the first view library is located, wherein the first subscription request is used for subscribing the first image and the first matching operation data.

For example, three algorithms a, B, and C are used for the first operation data corresponding to the first image in the current-level view library corresponding to the algorithm server, and the algorithm server runs the algorithm a, so that the portion determined by using the algorithm a in the first operation data is available for the algorithm server to perform image retrieval comparison, and the portion determined by using the algorithm a in the first operation data is the first matching operation data. At this time, a first subscription request is sent to a first server where the first view library is located, and the first subscription request is used for subscribing the first image and the first matching operation data. In this way, the algorithm server directly obtains the available operation data without performing calculation again according to the image to obtain the available operation data.

For example, as shown in FIG. 4, 5 view libraries are shown in FIG. 4 grouped into three levels, including a primary view library 410; the primary view library 410 corresponds to three secondary view libraries, which are a secondary view library 411, a secondary view library 412 and a secondary view library 413; the secondary view library 412 corresponds to the tertiary view library 4121. The algorithm server corresponding to the primary view library 410 is provided by an algorithm manufacturer 420 and an algorithm manufacturer 430; the algorithm server corresponding to the secondary view library 411 is provided by the algorithm manufacturer 430; the algorithm servers corresponding to the secondary view library 412, the secondary view library 413 and the tertiary view library 4121 are provided by the algorithm vendor 440. The algorithm vendor 420 may correspond to a first algorithm, the algorithm vendor 430 may correspond to a second algorithm, and the algorithm vendor 440 may correspond to a third algorithm.

For the primary view library 410, images may be subscribed to the secondary view library 411, the secondary view library 412, and the secondary view library 413. Secondary view library 412 may subscribe to tertiary view library 4121 for images;

for the first 430 algorithm server provided by the algorithm vendor 430, the current level view library corresponding to the first 430 algorithm server, i.e. the primary view library 410, may be subscribed, for the second 430 algorithm server provided by the algorithm vendor 430, the current level view library corresponding to the second 430 algorithm server, i.e. the secondary view library 411, may be subscribed, since a portion of the images in the primary view library 410 are from the secondary view library 411, the algorithm used in the secondary view library 411 is the second algorithm, therefore, the operation data of a part of images in the primary view library 410 is determined by using the second algorithm, and when the first 430 algorithm server provided by the algorithm manufacturer 430 subscribes to the primary view library 410, the operation data determined by the second algorithm is available to the first 430 algorithm server provided by the algorithm vendor 430, so that the corresponding operation data can be subscribed at the same time.

According to the embodiment of the invention, when the locally required operation data exists in the first view library to be subscribed, the first view library can be subscribed together, so that the resource consumption of local transportation can be reduced, and the resource utilization rate is improved.

After the subscription is successful, the view library server can send the updated data to the server provided by the algorithm manufacturer according to the synchronization frequency agreed by the subscription after determining the new subscribed data. Based on this, the method further comprises: and receiving gallery update data sent by the first server in response to the first subscription request, wherein the gallery update data comprises the updated first image, the updated first matching operation data and attribute information of an algorithm adopted by the first matching operation data.

In some embodiments, the method illustrated in FIG. 2 may also be applied to a view library server. For example, the upper view library may subscribe to the lower view library, and when receiving the data reported by the lower view library, it may be determined whether to accept the operation data uploaded by the lower view library as required; for example, before the upper view library sends a subscription request to the lower view library, it may be determined whether to accept the operation data of the lower view library as necessary. At this time, the first view library in the method shown in fig. 2 is a lower view library, and the method shown in fig. 2 may be applied to an upper view library server corresponding to the first view library.

For example, when the higher-level view library receives the face data of the lower-level view library, whether the feature value is needed by the current-level algorithm can be judged according to the requirement, and if the feature is needed, the feature needs to be subscribed when the lower-level subscription is performed.

In some embodiments, when the method is applied to a server provided by an algorithm vendor, the server provided by the algorithm vendor may subscribe to its corresponding current-level view library server respectively as needed, for example, may subscribe to the city of land. For example, when a server provided by an algorithm vendor initiates a subscription to the top-level view library server, the server provided by the algorithm vendor knows that the top-level view library includes 5 cities, and the server provided by the algorithm vendor may initiate 5 subscriptions to be respectively subscribed to each city. Thus, for algorithm manufacturers, some subscriptions can subscribe to the operation data, and some subscriptions only have pictures.

In some embodiments, if the operation data corresponding to the first image does not match the local requirement, only the image may be subscribed to, and at this time, the operation may be performed based on the image to obtain the desired operation data. Based on this, the method may further comprise the steps of:

step 1), if first matching operation data matched with an algorithm provided by a manufacturer does not exist in the first operation data, sending a second subscription request to a first server, wherein the second subscription request is used for subscribing a first image;

step 2), receiving gallery update data sent by the first server in response to the second subscription request, wherein the gallery update data comprises an updated first image;

and 3) calculating the updated first image based on an algorithm provided by a manufacturer to obtain first target calculation data corresponding to the updated first image.

The first server may be any one of the multilevel view library servers.

For the step 3), the method may be applied to a higher level view library server corresponding to the first view library or a server provided by a current level algorithm vendor. When the method is applied to the upper view library server corresponding to the first view library, the upper view library server may perform step 3) through a server provided by its corresponding algorithm vendor.

In some embodiments, if there is a first image without operation data, only the image may be subscribed to, and operation may be performed based on the image to obtain desired operation data. Based on this, the method may further comprise the steps of:

step a), determining whether a first image in a first view library to be subscribed corresponds to first operation data;

step b), if the first image does not correspond to the first operation data, sending a second subscription request to the first server, wherein the second subscription request is used for subscribing the first image;

step c), receiving gallery update data sent by the first server in response to the second subscription request, wherein the gallery update data comprises an updated first image;

and d), calculating the updated first image based on an algorithm provided by a manufacturer to obtain first target operation data corresponding to the updated first image.

For the step d), the method may be applied to a higher level view library server corresponding to the first view library or a server provided by a current level algorithm vendor. When the method is applied to the upper view library server corresponding to the first view library, the upper view library server may execute step d) through a server provided by a corresponding algorithm vendor.

In some embodiments, if the operation data corresponding to the first image does not match the local requirement, the method further comprises the steps of: and sending the first target operational data to the first server so that the first server updates the first operational data corresponding to the updated first image in the first view library based on the first target operational data. At this time, the types of the operation data of the first server can be enriched.

In some embodiments, after determining the updated view library data, the parsed updated view library data may be sent to a corresponding view library server for updating by the view library server. Based on this, the method further comprises: and sending the updated first image, the first operation data corresponding to the updated first image and the attribute information of the algorithm adopted by the first operation data to the second server, so that the second server can update the current-level view library based on the updated first image, the first operation data corresponding to the updated first image and the attribute information of the algorithm adopted by the first operation data.

In some embodiments, the first operation data corresponding to the first image is determined using an algorithm. As shown in fig. 5, image data is transmitted across the network in two view libraries at the same level. The picture source of the first network is a video stream that accesses an algorithm server provided by an algorithm vendor: the first video stream accesses the server provided by algorithm vendor A, the second video stream accesses the server provided by algorithm vendor B, because the same video stream is only extracted by the server provided by one algorithm vendor and the attribute identification, the relation between the picture and the feature/attribute in the view library of the first network is 1: 1. Accessing a video stream by an algorithm manufacturer, and extracting 'face picture + characteristic value (algorithm version + manufacturer) + face attribute' to a view library; each algorithm extracts a picture from the video stream to generate a unique id of the picture; and when the algorithm server corresponding to the view base server of the second network receives the cross-network picture, judging whether the algorithm version + manufacturer is in accordance with the local, if so, accessing the characteristic value + face attribute together, otherwise, accessing the picture only (which can be realized by different types of subscription). If the characteristic value and the face attribute are accessed, the algorithm server is directly used without secondary extraction; otherwise, only accessing the picture, and extracting the characteristic value and the face attribute by using an algorithm server.

As another example, the first operation data corresponding to the first image is determined using a plurality of algorithms. As shown in fig. 6, the upper view library collects the picture resources from the lower view library, and the algorithm vendor acquires the picture resources from the upper view library. The relationship between the picture and the feature/feature of the upper level gallery is 1: n, namely one picture has a plurality of first operation data determined corresponding to a plurality of algorithms.

And an algorithm server corresponding to the upper level view library in the first network subscribes data from the upper level view library in the first network, subscribes according to different cities in a similar cross-level upper level algorithm mode, if the operation data uploaded by a certain city is the manufacturer algorithm version of the factory, the user needs to subscribe the picture + the characteristics/attributes during subscription, otherwise, the user only needs to subscribe the picture. After subscription, the algorithm server corresponding to the upper-level view library can process the whole amount of human face image resources. I.e. algorithm servers of a plurality of different algorithm vendors all subscribe to the same picture.

After the algorithm manufacturer parses (there may be some data that does not need to be parsed because the subscribed data already has features/attributes), the parsed data (picture + features + face attributes) is pushed to the peer view library of the second network. Such that one picture corresponds to a plurality of features.

After receiving the peer view library of the second network, updating the storage of the peer view library, wherein the relationship between the pictures and the characteristics is 1: n is the same as the formula (I).

Fig. 7 is a flowchart illustrating an image data transmission method according to an embodiment of the present invention. As shown in fig. 7, the method may include the steps of:

and S710, determining attribute information of at least one algorithm adopted by second operation data corresponding to a second image in a second view library to be subscribed, wherein the second view library comprises the second image and the second operation data.

The method can be applied to a third server, wherein the third server is a server where the superior view library is located. The lower level servers corresponding to the third server comprise a plurality of lower level servers, each lower level server corresponds to one lower level view library, the second server is one or more of the lower level servers, and the second view library is one or more of the lower level view libraries.

Continuing with the example described in fig. 4, the third server may be the server in which the primary view library 410 is located, the second server may be any one of the server in which the secondary view library 411 is located, the server in which the secondary view library 412 is located, and the server in which the secondary view library 413 is located; or the third server may be the server where the secondary view library 412 resides and the second server may be the server where the tertiary view library 4121 resides. The attribute information of the algorithm can comprise an algorithm manufacturer identifier and/or an algorithm version identifier; the operational data may include image features and/or image attributes.

The image feature refers to a feature obtained by performing feature extraction on the first image according to an algorithm. The image attribute may refer to a result of identifying the first image based on the extracted image feature. For example, the gender recognition of the human face in the first image can be performed based on the extracted image features, resulting in a gender recognition result, which can be considered as an image attribute.

As one example, the multi-level view library server may include a lower level view library server and an upper level view library server. The feature values of the algorithms and the algorithm versions can be collected together when the face pictures are collected by the lower level image libraries. When the superior view library summarizes resources, the pictures, the features and the algorithm version are summarized together. When subscribing to the upper level view library, the upper level algorithm manufacturer can perform classified and local subscription, and subscribe 'picture + feature' if the algorithm feature exists; otherwise, only subscribing to the picture.

As another example, the multi-level view library server may also belong to different networks, and peer cross-network transmission directly synchronizes the eigenvalue of the algorithm and the algorithm version to the view library together, and the view library performs cross-network; for example, when a server in a first network subscribes to data from other networks, the server in the first network can subscribe according to the needs of the server.

For the lower view library server, according to the standard, when the face information is collected in the gallery, the face features can be uploaded, and when the face is applied to uploading the face attributes and pictures, the features of the face small picture and the algorithm version and the like can be uploaded together.

S720, determining whether second matching operation data matched with the view base algorithm exists in the second operation data based on the attribute information of at least one algorithm adopted by the second operation data;

if the second operation data corresponding to the second image is multiple, the current second operation data can be selected from the multiple second operation data corresponding to the second image in sequence for matching until the matching is successful or the multiple second operation data corresponding to the first image is traversed.

As an example, when receiving the reported data from the lower view library, the upper view library may determine whether to receive the operation data from the lower view library as needed. For example, when receiving the face data synchronized with the lower view library, the upper view library may determine whether the face data includes a feature value required by the current level, and if necessary, may subscribe to the feature when subscribing to the lower level. Otherwise, directly subscribing the picture.

S730, if second matching operation data matched with the view library algorithm exists in the second operation data, sending a third subscription request to a second server where a second view library is located, wherein the third subscription request is used for subscribing a second image and the second matching operation data; and receiving gallery update data sent by the second server in response to the third subscription request, wherein the gallery update data comprises the updated second image, second matching operation data corresponding to the updated second image and attribute information of an algorithm adopted by the second matching operation data.

And S740, updating the current-level view library based on the update data of the view library.

In some embodiments, if there is no second matching operation data in the second operation data that matches the view library algorithm, then only the image may be subscribed to, based on which the method further comprises:

step A), if second matching operation data matched with the view library algorithm does not exist in the second operation data, sending a fourth subscription request to a second server where a second view library is located, wherein the fourth subscription request is used for subscribing to a second image; receiving gallery update data sent by the second server in response to the fourth subscription request, wherein the gallery update data comprises an updated second image;

step B), the updated second image is sent to a server provided by an algorithm manufacturer, so that the server provided by the algorithm manufacturer can calculate the updated second image based on a manufacturer algorithm matched with the view base algorithm, and second target operation data corresponding to the updated second image are obtained;

step C), receiving second target operation data from a server provided by an algorithm manufacturer and attribute information of the manufacturer algorithm;

and D), updating second operation data corresponding to the updated second image in the second view library based on the second target operation data.

According to the embodiment of the application, when the locally required operation data exists in the second view library to be subscribed, the subscription can be carried out together, so that the resource consumption of local transportation can be reduced, and the resource utilization rate is improved.

Fig. 8 is a schematic structural diagram of an image data transmission device according to an embodiment of the present invention. As shown in fig. 8, the apparatus may include:

a determining module 801, configured to determine attribute information of at least one algorithm used in first operation data corresponding to a first image in a first view library to be subscribed, where the first view library includes the first image and the first operation data;

a matching module 802, configured to determine, based on attribute information of at least one algorithm used by the first operation data, whether there is first matching operation data matching an algorithm provided by a manufacturer in the first operation data;

the subscription module 803, if the first operational data includes first matching operational data matching the algorithm provided by the vendor, sends a first subscription request to the first server where the first view library is located, where the first subscription request is used to subscribe the first image and the first matching operational data.

In some embodiments, further comprising:

the subscription module 803 is further configured to send a second subscription request to the first server if there is no first matching operation data matching the algorithm provided by the vendor in the first operation data, where the second subscription request is used to subscribe to the first image;

the receiving module is used for receiving the gallery update data sent by the first server in response to the second subscription request, and the gallery update data comprises an updated first image;

and the operation module is used for operating the updated first image based on an algorithm provided by a manufacturer to obtain first target operation data corresponding to the updated first image.

In some embodiments, further comprising:

the determining module 801 is further configured to determine whether a first image in the first view library to be subscribed corresponds to first operation data;

the subscription module 803 is further configured to, if the first image does not correspond to the first operation data, send a second subscription request to the first server, where the second subscription request is used to subscribe to the first image;

the receiving module is used for receiving gallery updating data sent by the first server in response to the second subscription request, and the gallery updating data comprises an updated first image;

and the updating module is used for calculating the updated first image based on an algorithm provided by a manufacturer to obtain first target operation data corresponding to the updated first image.

In some embodiments, further comprising:

the sending module is used for sending the first target operational data to the first server so that the first server updates the first operational data corresponding to the updated first image in the first view library based on the first target operational data.

In some embodiments, the attribute information of the algorithm includes an algorithm vendor identification and an algorithm version identification; the operational data includes image features and/or image attributes.

Fig. 9 is a schematic structural diagram of an image data transmission device according to an embodiment of the present invention. As shown in fig. 9, the apparatus may include:

a determining module 901, configured to determine attribute information of at least one algorithm used in second operation data corresponding to a second image in a second view library to be subscribed, where the second view library includes the second image and the second operation data;

a matching module 902, configured to determine whether second matching operation data matching the view library algorithm exists in the second operation data based on attribute information of at least one algorithm used by the second operation data;

a subscription module 903, configured to send a third subscription request to a second server where a second view library is located if second matching operation data matching the view library algorithm exists in the second operation data, where the third subscription request is used to subscribe a second image and the second matching operation data; receiving gallery updating data sent by the second server in response to the third subscription request, wherein the gallery updating data comprises an updated second image, second matching operation data corresponding to the updated second image and attribute information of an algorithm adopted by the second matching operation data;

and an updating module 904 for updating the current level view base based on the gallery update data.

In some embodiments, the apparatus is applied to a third server, where the third server includes a plurality of lower level servers, each lower level server corresponds to a lower level view library, the second server is one or more of the plurality of lower level servers, and the second view library is one or more of the lower level view libraries.

In some embodiments, further comprising:

the subscription module 903 is further configured to send a fourth subscription request to a second server where the second view library is located if second matching operation data matching the view library algorithm does not exist in the second operation data, where the fourth subscription request is used to subscribe to the second image; receiving gallery update data sent by the second server in response to the fourth subscription request, wherein the gallery update data comprises an updated second image;

the receiving module is used for receiving second target operation data from a server provided by an algorithm manufacturer and attribute information of the manufacturer algorithm;

the updating module 904 is further configured to update second operation data corresponding to the updated second image in the second view library based on the second target operation data.

The device provided by the embodiment of the present invention has the same implementation principle and technical effect as the method embodiments, and for the sake of brief description, reference may be made to the corresponding contents in the method embodiments without reference to the device embodiments.

Further, the present embodiment also provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to perform the steps of the method provided by the foregoing method embodiment.

The computer program product of the face recognition method, the face recognition device and the face recognition system provided by the embodiment of the invention comprises a computer readable storage medium storing a program code, wherein instructions included in the program code can be used for executing the method in the foregoing method embodiment, and specific implementation can refer to the method embodiment, which is not described herein again.

It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, a division of a unit is merely a division of one logic function, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.

Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will understand that: those skilled in the art can still make modifications or changes to the embodiments described in the foregoing embodiments, or make equivalent substitutions for some features, within the scope of the disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein.

Claims

1. An image data transmission method, comprising:

determining whether first matching operation data matched with an algorithm provided by a manufacturer exists in the first operation data based on attribute information of at least one algorithm adopted by the first operation data;

if first matching operation data matched with an algorithm provided by a manufacturer exist in the first operation data, sending a first subscription request to a first server where the first view library is located, wherein the first subscription request is used for subscribing the first image and the first matching operation data.

2. The method of claim 1, further comprising:

3. The method according to claim 1, further comprising, before the determining attribute information of at least one algorithm used for the first operation data corresponding to the first image in the first view library to be subscribed, the following:

determining whether a first image in a first view library to be subscribed corresponds to first operation data;

4. The method of claim 2 or 3, further comprising:

5. The method of claim 1, wherein the attribute information of the algorithm includes an algorithm vendor identifier and an algorithm version identifier; the operational data includes image features and/or image attributes.

6. An image data transmission method, comprising:

and updating the current-level view library based on the gallery update data.

7. The method as claimed in claim 6, wherein the third server is applied to a plurality of lower level servers, each of the lower level servers corresponds to a lower level view library, the second server is one or more of the lower level servers, and the second view library is one or more of the lower level view libraries.

8. The method of claim 6, further comprising:

9. An image data transmission apparatus, characterized by comprising:

10. An image data transmission apparatus, characterized by comprising:

the subscription module is used for sending a third subscription request to a second server where the second view library is located if second matching operation data matched with the view library algorithm exists in the second operation data, wherein the third subscription request is used for subscribing the second image and the second matching operation data; receiving gallery update data sent by the second server in response to the third subscription request, wherein the gallery update data comprises an updated second image, second matching operation data corresponding to the updated second image and attribute information of an algorithm adopted by the second matching operation data;

and the updating module is used for updating the current-level view library based on the map library updating data.

11. An electronic device, comprising: a processing device and a storage device;

the storage means has stored thereon a computer program which, when executed by the processing means, performs the method of any one of claims 1 to 8 for image data transmission.

12. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of the preceding claims 1 to 8.