CN109522789A

CN109522789A - Eyeball tracking method, apparatus and system applied to terminal device

Info

Publication number: CN109522789A
Application number: CN201811163385.8A
Authority: CN
Inventors: 孔祥晖; 严海; 黄通兵
Original assignee: Beijing Qixin Yiwei Information Technology Co Ltd
Current assignee: Beijing 7Invensun Technology Co Ltd; Beijing Qixin Yiwei Information Technology Co Ltd
Priority date: 2018-09-30
Filing date: 2018-09-30
Publication date: 2019-03-26
Also published as: WO2020063021A1

Abstract

The invention discloses a kind of eyeball tracking method, apparatus and system applied to terminal device.Wherein, this method comprises: acquisition image information；Eye feature information is extracted from image information, forms eye feature information aggregate, wherein includes eye feature information in image information；Eye feature information aggregate is sent to server；Receive and parse through server feedback watches information attentively.The eyeball tracking algorithm that the present invention solves the prior art runs the technical problem for causing the resource consumption of equipment end big in equipment end.

Description

Eyeball tracking method, apparatus and system applied to terminal device

Technical field

The present invention relates to eyeball tracking technical fields, in particular to a kind of eyeball tracking applied to terminal device Method, apparatus and system.

Background technique

Interactive mode of the eyeball tracking technology as innovation, it is more and more by big well known, in people work, study Etc. be widely used.Wherein, the process flow of eyeball tracking technology mainly includes following three step:

Step S1, eyeball tracking equipment obtain the face-image of user using acquisition equipment, wherein acquiring equipment can be Optical devices, electrical devices etc.；

Step S2, eyeball tracking equipment handle the face-image of user, and extract the eye feature information of user；

Step S3, eyeball tracking equipment handle the eye feature information of user, seek the direction of gaze of user with And blinkpunkt position.

Currently, above three step is realized in equipment end.However, realizing the mistake to the eyeball tracking of user in equipment end The consumption of Cheng Zhong, the resources such as the system service ability of equipment end, memory space are bigger, in addition, if long-play eye The program of ball tracking, equipment end are likely to occur the problems such as fever is hot, and power consumption increases, and the above problem reduces the use of user Experience.

Cause the resource consumption of equipment end big in equipment end operation for the eyeball tracking algorithm of the above-mentioned prior art Problem, currently no effective solution has been proposed.

Summary of the invention

The embodiment of the invention provides a kind of eyeball tracking method, apparatus and system applied to terminal device, so that The eyeball tracking algorithm for solving the prior art less runs the technical problem for causing the resource consumption of equipment end big in equipment end.

According to an aspect of an embodiment of the present invention, a kind of eyeball tracking method applied to terminal device is provided, is wrapped It includes: acquisition image information；Image information is sent to cloud processor；Receive and parse through cloud processor feedback watches information attentively.

Further, applied to the eyeball tracking method of terminal device further include: obtain and send information；It determines and sends information Corresponding image identification, wherein image identification is used to identify the transmission sequence of first frame image；Information will be sent with the first rule Mode is sent to cloud processor.

Further, applied to the eyeball tracking method of terminal device further include: determine the first frame image for sending information Corresponding timestamp；The corresponding image identification of first frame image is generated according to timestamp.

Further, sending information includes: the first frame head, the first control word, the first check code and cloud information.

Further, applied to the eyeball tracking method of terminal device further include: what reception was sent in a manner of Second Rule Receive information；Parsing receives information, receives information and includes the second frame head, the second control word and the second check code and watch letter attentively Breath；Completion verifies the second check code；It is true by information is watched attentively if the check results of the second check code meet preset condition Be set to target object watches information attentively.

According to another aspect of an embodiment of the present invention, a kind of eyeball tracking method applied to terminal device is additionally provided, It include: the image information for obtaining equipment end and sending；Eye feature information is extracted from image information, forms eye feature information collection It closes, wherein include eye feature information in image information；Information is watched attentively according to the determination of eye feature information aggregate；Transmission is watched attentively Information is to equipment end.

According to another aspect of an embodiment of the present invention, a kind of eyeball tracking method applied to terminal device is additionally provided, It include: acquisition image information；Eye feature information is extracted from image information, forms eye feature information aggregate, wherein image It include eye feature information in information；Eye feature information aggregate is sent to server；Receive and parse through the note of server feedback Visual information.

Further, before forming eye feature information aggregate, the eyeball tracking method applied to terminal device is also wrapped It includes: converting corresponding eye feature information for image information；By corresponding eye feature asset data label；According to finger The corresponding eye feature information of configured data label is completed sequence by fixed sequence.

Further, corresponding eye feature information include the pupil information of target object, cornea information, facula information, Iris information and/or eyelid information, eyes image information one or more of watch depth information attentively.

Further, eye feature information aggregate is configured as one or more groups of eye features letter with data label Breath.

Further, applied to the eyeball tracking method of terminal device further include: obtain and send information；It determines and sends information Corresponding image identification obtains the eye feature information aggregate of image identification corresponding series, wherein image identification is for identifying the The transmission sequence of one frame image；Information will be sent, server is sent to the first regular fashion.

Further, information is sent further include: the first frame head, the first control word and the first check code.

According to another aspect of an embodiment of the present invention, a kind of eyeball tracking device applied to terminal device is additionally provided, It include: the first acquisition module, for acquiring image information；First sending module, for sending image information to cloud processor； First receiving module, for receive and parse through cloud processor feedback watch information attentively.

According to another aspect of an embodiment of the present invention, a kind of eyeball tracking device applied to terminal device is additionally provided, It include: the second acquisition module, for acquiring image information；Second extraction module, for extracting eye feature from image information Information forms eye feature information aggregate, wherein includes eye feature information in image information；Second sending module, for sending out Send eye feature information aggregate to server；Second receiving module watches information attentively for receive and parse through server feedback.

According to another aspect of an embodiment of the present invention, a kind of eyeball tracking system applied to terminal device is additionally provided, Include: equipment end, for acquiring image information, and extracts eye feature information from image information, form eye feature information Set, is then sent to server for eye feature information, receive and parse through server feedback watches information attentively；Server is used In the eye feature information that receiving device end is sent, and eye feature information is handled, obtain target object watches letter attentively Breath.

According to another aspect of an embodiment of the present invention, a kind of storage medium is additionally provided, which includes storage Program, wherein program executes the eyeball tracking method for being applied to terminal device.

According to another aspect of an embodiment of the present invention, a kind of processor is additionally provided, which is used to run program, In, the eyeball tracking method for being applied to terminal device is executed when program is run.

In embodiments of the present invention, using equipment end in such a way that server combines, image information is acquired in equipment end, And eye feature information is extracted from image information, eye feature information aggregate is formed, then sends out eye feature information aggregate It send to server, is handled by server according to eye feature information aggregate, obtain the information of watching attentively of target object, and by mesh The information of watching attentively of mark object is sent to equipment end, so equipment end receive and parse through server feedback watch information attentively.

In above process, the acquisition of the image information of target object, the extraction of eye feature information aggregate are in equipment end Operation, and the treatment process of eye feature information aggregate is realized in the server.Due to being carried out according to eye feature information aggregate The resource that processing obtains watching attentively the blinkpunkt algorithm consumption of information is more, therefore, it is handled in server, is effectively reduced The resource consumption of equipment end.In addition, causing resource if image, which is directly uploaded to server, will occupy more bandwidth Consumption, therefore, in this application, equipment end can extract eye feature information aggregate from image information, and by eye feature Information aggregate is sent to server, rather than image information is sent directly to server, can reduce image information directly It is uploaded to the occupied bandwidth of server, further decreases the resource consumption of equipment end.

It can be seen that the eyeball tracking algorithm that scheme provided herein can solve the prior art is transported in equipment end The technical problem that row causes the resource consumption of equipment end big.

Detailed description of the invention

The drawings described herein are used to provide a further understanding of the present invention, constitutes part of this application, this hair Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:

Fig. 1 is a kind of flow chart of eyeball tracking method applied to terminal device according to an embodiment of the present invention；

Fig. 2 is a kind of schematic diagram of optional frame format according to an embodiment of the present invention；

Fig. 3 is a kind of flow chart of eyeball tracking method applied to terminal device according to an embodiment of the present invention；

Fig. 4 is a kind of structural schematic diagram of eyeball tracking device applied to terminal device according to an embodiment of the present invention；

Fig. 5 is a kind of structural schematic diagram of eyeball tracking device applied to terminal device according to an embodiment of the present invention；

Fig. 6 is a kind of structural schematic diagram of eyeball tracking system applied to terminal device according to an embodiment of the present invention；

Fig. 7 is a kind of flow chart of eyeball tracking method applied to terminal device according to an embodiment of the present invention；

Fig. 8 is a kind of flow chart of eyeball tracking method applied to terminal device according to an embodiment of the present invention；And

Fig. 9 is a kind of flow chart of eyeball tracking method applied to terminal device according to an embodiment of the present invention.

Specific embodiment

In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people The model that the present invention protects all should belong in member's every other embodiment obtained without making creative work It encloses.

It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, " Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way Data be interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to illustrating herein or Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product Or other step or units that equipment is intrinsic.

Embodiment 1

According to embodiments of the present invention, a kind of eyeball tracking embodiment of the method applied to terminal device is provided, is needed Bright, step shown in the flowchart of the accompanying drawings can be held in a computer system such as a set of computer executable instructions Row, although also, logical order is shown in flow charts, and it in some cases, can be to be different from sequence herein Execute shown or described step.

In addition it is also necessary to explanation, the eyeball for being applied to terminal device provided by the present embodiment is can be performed in equipment end Method for tracing.Wherein, equipment end includes movable equipment and irremovable equipment, and movable equipment can be but be not limited to hand The equipment such as machine, plate.

Fig. 1 is the flow chart of the eyeball tracking method according to an embodiment of the present invention applied to terminal device, such as Fig. 1 institute Show, this method comprises the following steps:

Step S102 acquires image information.

It should be noted that equipment end has image acquisition device, which can acquire the image letter of target object Breath, wherein include the eyes image of target object in the image information of target object.Optionally, in the feelings that equipment end is mobile phone Under condition, image acquisition device can be the camera of mobile phone.Before the resolution ratio of the rear camera of usual mobile phone is higher than mobile phone The resolution ratio of camera is set, but the resolution ratio of the front camera of mobile phone is millions pixel, acquired image information is full The demand of sufficient eyeball tracking technology.In the case where mobile phone has front camera, the image acquisition device in the present embodiment is selected The front camera of mobile phone.

Step S104 sends image information to cloud processor.

It should be noted that the processing capacity due to cloud server is stronger, more complicated algorithm can be run, because This, after collecting the image information of target object, the image information of target object is sent to cloud processor by equipment end It is handled, to achieve the purpose that reduce equipment end resource consumption.Wherein, cloud processor is the processing in cloud server Device.

In addition, equipment end can be communicated through wireless communication with cloud processor, wherein communication It can include but is not limited to the modes such as WIFI, 3G, 4G, GPRS.

In a kind of optional scheme, after receiving image information, cloud processor is mentioned based on data extraction algorithm Take at least one of information in image information: pupil information, cornea information, iris information and/or eyelid information, eye figure As information, watch depth information attentively, and at least one general information is as eye feature information.After obtaining eye feature information, Cloud processor watches information attentively according to eyeball tracking algorithm calculating target object, wherein watches information attentively including at least the side of watching attentively To information and blinkpunkt information, wherein blinkpunkt information can be seat of the blinkpunkt of target object on equipment end screen Mark information.Then, the information of watching attentively being calculated is sent to equipment end, equipment by cloud processor through wireless communication End, which can receive, watches information attentively.

Step S106, receive and parse through cloud processor feedback watches information attentively.

It should be noted that equipment end can be communicated through wireless communication with cloud processor, wherein wireless Communication mode can include but is not limited to the modes such as WIFI, 3G, 4G, GPRS.

In addition it is also necessary to explanation, cloud processor is not necessarily to image information being sent to equipment end, will only watch information attentively It is sent to equipment end.The traffic rate of cloud processor and equipment end can satisfy requirement of real-time, and handling capacity compared with It is small, it can satisfy requirement using traffic rate slower GPRS.

It based on scheme defined by above-mentioned steps S102 to step S106, can know, be handled using equipment end and cloud Image information in the image information of equipment end acquisition target object, and is sent to cloud processor, cloud by the mode combined Processor extracts eye feature information after receiving image information from image information, forms eye feature information, then Eye feature information is handled, obtains the information of watching attentively of target object, and the information of watching attentively of target object is sent to and sets Standby end, so equipment end receive and parse through cloud processor feedback watch information attentively.

It is easily noted that, disappears due to being handled to obtain the blinkpunkt algorithm for watching information attentively according to eye feature information The resource of consumption is more, and therefore, by it, processor is handled beyond the clouds, effectively reduces the resource consumption of equipment end.In addition, by It is stronger in the processing capacity of cloud processor, more complicated algorithm can be run, therefore, cloud processor can be directly to target The image information of object is handled to obtain eye feature information, and according to eye feature information extraction to information is watched attentively, finally Information will be watched attentively and be sent to equipment end, further decrease the resource consumption of equipment end.

It should be noted that equipment end can be extracted from image information after collecting the image information of target object The eye feature information of series.Specifically, equipment end extracts at least one of in image information based on data extraction algorithm Information: pupil information, cornea information, facula information, iris information and/or eyelid information, eyes image information watch depth letter attentively Breath, and at least one general information is as eye feature information.

In addition it is also necessary to explanation, equipment end acquired image information may be to include the image in multiple image Information, in order to guarantee pupil information, cornea information, facula information, iris information and/or eyelid information, eyes image information, note It is the eye feature information aggregate of same width frame image zooming-out depending on depth information, image information is being sent to cloud processor In the process, image identification is carried out to current frame image.Detailed process may include steps of；

Step S1060 is obtained and is sent information；

Step S1062 is determined and is sent the corresponding image identification of information, wherein image identification, which is used to identify, sends information Transmission sequence；

Step S1064 will send information with the first regular fashion and be sent to processor.

It should be noted that image identification can be but be not limited to timestamp, title or be assigned randomly to mark.Its In, in the case where image identification is timestamp, which is to acquire timestamp corresponding to the time of image information.It is optional , above-mentioned transmission information can be information corresponding to current frame image, wherein equipment end can determine the first frame for sending information The corresponding timestamp of image, and the corresponding image identification of first frame image is generated according to timestamp.

It is easily noted that, in scheme defined by above-mentioned steps S1060 to step S1064, is sending image information To cloud processor, image identification is also sent to cloud processor.Cloud processor can be according to every frame image Image identification receives in sequence, ensure that correct parsing of the cloud processor to the image information after mark, and then to correct Eye feature information handled, obtain accurate target object watches information attentively.

In a kind of optional scheme, equipment end needs to configure image before sending image identification correspondence image information Then the corresponding transmission information of information will send information again with the first regular fashion and be sent to cloud processor.Wherein, letter is sent Breath includes: the first frame head, the first control word, the first check code and image information.Optionally, the first regular fashion can be frame Format, wherein Fig. 2 shows a kind of schematic diagrames of optional frame format, and as shown in Figure 2, frame format includes at least four parts, i.e., Frame head, control word, data (i.e. image information) and check code, wherein check code can be but be not limited by CRC (Cyclic Redundancy Check, cyclic redundancy check) verifies obtained number.

It should be noted that after can guaranteeing that equipment end is communicated with cloud processor using CRC check, cloud processor Correct data can be resolved to, and guarantee the correctness of data, while can also prevent influence of the interference signal to parsing result.

In addition it is also necessary to explanation, the processor of server is after receiving eye feature information, to eye feature Information is parsed, and is handled the data after parsing, obtains the information of watching attentively of target object, and according to Second Rule side Formula will watch information attentively and be sent to equipment end, at this point, the data in frame format are to watch information attentively.Equipment end is being received with Second Rule The reception information that mode is sent, and reception information is parsed, then complete the verification to the second check code.Wherein, if the second school The check results for testing code meet preset condition, then equipment end watches information attentively for watch that information is determined as target object attentively.Optionally, Information is received to include: the second frame head, the second control word and the second check code and watch information attentively.

It should be noted that equipment end is after the data for receiving Second Rule mode, using check code to parsing after Data carry out verification and can guarantee the correctness of data, while can also prevent influence of the interference signal to parsing result, thus Obtain correct target object watches information attentively.

As shown in the above, eyeball tracking technology can be effectively reduced to the meter of equipment end in scheme provided herein The requirement of calculation ability.In addition, the extraction to information is watched attentively is completed by cloud processor real-time update eyeball tracking algorithm, it can With the eyeball tracking processing speed at lifting means end.Finally, scheme provided herein is calculated in the processing of equipment end operation image Method runs blinkpunkt algorithm by cloud processor, can substantially reduce the handling capacity of communication data, improves traffic rate.

Embodiment 2

According to embodiments of the present invention, a kind of eyeball tracking embodiment of the method applied to terminal device is provided, is needed Bright, the eyeball tracking method for being applied to terminal device provided by the present embodiment can be performed in server.Wherein, Fig. 9 is root According to the flow chart of the eyeball tracking method applied to terminal device of the embodiment of the present invention, as shown in figure 9, this method includes as follows Step:

Step S902 acquires image information.

It should be noted that equipment end has image acquisition device, which can acquire the image letter of target object Breath, wherein include the eyes image of target object in the image information of target object.Optionally, in the feelings that equipment end is mobile phone Under condition, image acquisition device can be the camera of mobile phone.Before the resolution ratio of the rear camera of usual mobile phone is higher than mobile phone The resolution ratio of camera is set, but the resolution ratio of the front camera of mobile phone is millions pixel, acquired image information is full The demand of sufficient eyeball tracking technology.In the case where mobile phone has front camera, image acquisition device in the present embodiment is selected With the front camera of mobile phone.

Step S904 extracts eye feature information from image information, forms eye feature information aggregate, wherein image It include eye feature information in information.

It should be noted that eye feature information aggregate is configured as one or more groups of eye features with data label Information.In addition, equipment end also converts the image information of target object to corresponding before forming eye feature information aggregate Eye feature information, and by corresponding eye feature asset data label, then according to specified sequence by configured data The corresponding eye feature information of label completes sequence.Wherein, corresponding eye feature information includes the pupil letter of target object Breath, cornea information, facula information, iris information and/or eyelid information, eyes image information one of watch depth information attentively It is or several, wherein pupil information includes at least: pupil center location, pupil diameter；Cornea information includes at least: cornea hot spot Reflective information；Iris information includes at least: iris edge information.

Step S906 sends eye feature information aggregate to server.

It should be noted that the server in step S906 can be cloud processor, wherein equipment end and server can It is communicated through wireless communication, wherein communication can include but is not limited to WIFI, 3G, 4G, GPRS etc. Mode.

In addition it is also necessary to explanation, during equipment end sends eye feature information aggregate to server, equipment Eye feature information is only sent to server it is not necessary that collected whole image information is sent to server by end.It can Choosing, equipment end can send eye feature information to server according to the form of frame format, wherein frame format can be by frame Head, control word, human eye data, check code constitute, equipment end equipment end send eye feature information to server during, It can be according to 1 byte frame head, 1 byte control word, the eye feature information of 3 to 6 bytes (i.e. human eye data), 2 byte CRC checks (i.e. check code) is sent to server, wherein the quantity total amount of transmission is about 10 bytes.In addition, in 90 frame images of processing per second In the case where data, the data volume that equipment end is sent to server is 90*10*8=7.2kbps.It can be seen that equipment end and clothes The traffic rate of business device can satisfy requirement of real-time, and handling capacity is smaller, even if the GPRS slower using traffic rate It can satisfy requirement.

In a kind of optional scheme, equipment end collects the two-dimension human face image of target object, wherein two-dimension human face figure The resolution ratio of picture is 1920*1080.Equipment end handles two-dimension human face image, obtains the filled circles of 6 pixels, wherein The centre coordinate of the filled circles is (800*800,840*800).At this point, equipment end passes through the centre coordinate of the filled circles wirelessly The mode of communication is sent to server.

Step S908, receive and parse through server feedback watches information attentively.

In a kind of optional scheme, server is after receiving eye feature information aggregate, and server is according to eyeball Tracing algorithm calculating target object watches information attentively, wherein watches information attentively and believes including at least direction of gaze information and blinkpunkt Breath, wherein blinkpunkt information can be coordinate information of the blinkpunkt of target object on equipment end screen.Then, server The information of watching attentively being calculated is sent to equipment end through wireless communication, information is watched in equipment end reception attentively.It needs Bright, in above process, server is not necessarily to image information being sent to equipment end, will only watch information attentively and be sent to equipment end ?.Server and the traffic rate of equipment end can satisfy requirement of real-time, and handling capacity is smaller, even if using communication speed The slower GPRS of rate also can satisfy requirement.

In addition it is also necessary to explanation, since computing capability of the eyeball tracking algorithm to equipment end is more demanding, eyeball is chased after Track algorithm is stringent to the kernel of equipment end and the requirement of dominant frequency, and the equipment end of middle-end level is difficult to meet the hardware index.And It is larger to the resource consumption of equipment end and if running eyeball tracking algorithm always in equipment end, even result in equipment end Fever, Caton.And scheme provided herein runs eyeball tracking algorithm in server, it is possible to prevente effectively from above-mentioned ask Topic.

In addition, the limited battery capacity of equipment end, and eyeball tracking algorithm is a technology based on video flowing, power consumption Amount is higher, therefore, eyeball tracking algorithm is run in server, can reduce the energy consumption of equipment end.

Based on scheme defined by above-mentioned steps S902 to step S908, it can know, using equipment end and server phase In conjunction with mode, equipment end acquisition target object image information, and from image information extract eye feature information, formed Then eye feature information aggregate is sent to server by eye feature information aggregate, by server according to eye feature information Set is handled, and obtains the information of watching attentively of target object, and the information of watching attentively of target object is sent to equipment end, Jin Ershe Standby end receives and what resolution server was fed back watches information attentively.

It is easily noted that, the acquisition of the image information of target object, the extraction of eye feature information aggregate are in equipment End operation, and the treatment process of eye feature information aggregate is realized in the server.Due to according to eye feature information aggregate into The resource that row processing obtains watching attentively the blinkpunkt algorithm consumption of information is more, therefore, it is handled in server, is effectively dropped The low resource consumption of equipment end.In addition, causing to provide if image, which is directly uploaded to server, will occupy more bandwidth Source consumption, therefore, because the processing capacity of server is stronger, can run more complicated algorithm, therefore, in this application, Equipment end can also extract eye feature information aggregate from image information, and eye feature information aggregate is sent to service Device, rather than image information is sent directly to server, it can reduce and directly image information is uploaded to occupied by server Bandwidth, further decrease the resource consumption of equipment end.

It should be noted that equipment end can be extracted from image information after collecting the image information of target object Eye feature information.Specifically, equipment end extracts at least one of information in image information: pupil based on data extraction algorithm Hole information, cornea information, facula information, iris information and/or eyelid information, eyes image information watch depth information attentively, and will At least one information as eye feature information.

In addition it is also necessary to explanation, the eye feature information aggregate extracted from image information may include multiframe Eye feature information in image, in order to guarantee pupil information, cornea information, facula information, iris information and/or eyelid letter Breath, watches the eye feature information that depth information is same width frame image zooming-out attentively at eyes image information, by eye feature information During set is sent to server, image identification is carried out to current frame image.Detailed process may include steps of；

Step S2060 is obtained and is sent information；

Step S2062 is determined and is sent the corresponding image identification of information, obtains the eye feature letter of image identification corresponding series Breath set, wherein image identification is used to identify the transmission sequence for sending information；

Step S2064 will send information with the first regular fashion and be sent to server.

It is easily noted that, in scheme defined by above-mentioned steps S2060 to step S2064, is sending eye feature During information aggregate to server, image identification is also sent to server.Server can be according to the image of every frame image Mark receives in sequence, ensure that correct parsing of the server to the eye feature information after mark, and then to correct eye Portion's information is handled, and obtain accurate target object watches information attentively.

In a kind of optional scheme, equipment end send image identification corresponding series eye feature information aggregate it Before, the corresponding transmission information of eye feature information is needed to configure, information will be sent again then with the first regular fashion and be sent to clothes Business device.Wherein, sending information includes: the first frame head, the first control word, the first check code and eye feature information aggregate.It can Choosing, the first regular fashion can be frame format, wherein Fig. 2 shows a kind of schematic diagrames of optional frame format, can by Fig. 2 Know, frame format includes at least four parts, i.e. frame head, control word, data (i.e. eye feature information) and check code, wherein school Testing code can be but be not limited by the number that CRC (Cyclic Redundancy Check, cyclic redundancy check) verification obtains Word.It should be noted that server can be resolved to correctly after can guaranteeing equipment end and server communication using CRC check Data, and guarantee the correctness of data, while can also prevent influence of the interference signal to parsing result.

In addition it is also necessary to explanation, server carries out eye feature information after receiving eye feature information Parsing, and the data after parsing are handled, the information of watching attentively of target object is obtained, and will watch attentively according to Second Rule mode Information is sent to equipment end, at this point, the data in frame format are to watch information attentively.Equipment end is sent in a manner of Second Rule in reception Reception information, and parse reception information, then complete the verification to the second check code.Wherein, if the school of the second check code It tests result and meets preset condition, then equipment end watches information attentively for watch that information is determined as target object attentively.Optionally, information is received Include: the second frame head, the second control word and the second check code and watches information attentively.It should be noted that equipment end is receiving To after the data of Second Rule mode, the correct of data can be guaranteed by being verified using check code to the data after parsing Property, while can also prevent influence of the interference signal to parsing result, so that obtain correct target object watches information attentively.

As shown in the above, eyeball tracking technology can be effectively reduced to the meter of equipment end in scheme provided herein The requirement of calculation ability.In addition, completing the extraction to information is watched attentively, Ke Yiti by server real-time update eyeball tracking algorithm Rise the eyeball tracking processing speed of equipment end.Finally, scheme provided herein is in equipment end operation image Processing Algorithm, by Server runs blinkpunkt algorithm, can substantially reduce the handling capacity of communication data, improves traffic rate.

Embodiment 3

According to embodiments of the present invention, a kind of eyeball tracking embodiment of the method applied to terminal device is provided, is needed Bright, the eyeball tracking method for being applied to terminal device provided by the present embodiment can be performed in server, and server can be Cloud Server.Wherein, Fig. 3 is the flow chart of the eyeball tracking method according to an embodiment of the present invention applied to terminal device, such as Shown in Fig. 3, this method comprises the following steps:

Step S302, the eye feature information for the target object that receiving device end is sent.

It should be noted that equipment end includes movable equipment and irremovable equipment, movable equipment can be but not It is limited to the equipment such as mobile phone, plate.Equipment end has image acquisition device, which can acquire the image letter of target object Breath, wherein include the eyes image of target object in the image information of target object.Optionally, in the feelings that equipment end is mobile phone Under condition, image acquisition device can be the camera of mobile phone.Before the resolution ratio of the rear camera of usual mobile phone is higher than mobile phone The resolution ratio of camera is set, but the resolution ratio of the front camera of mobile phone is millions pixel, acquired image information is full The demand of sufficient eyeball tracking technology.In the case where mobile phone has front camera, the image acquisition device of the present embodiment selects hand The front camera of machine.

In addition, the eye feature information of target object includes at least one following: pupil information, cornea information, hot spot letter Breath, eyelid information, eyes image information, watches depth information attentively at iris information, wherein pupil information includes at least: pupil center Position, pupil diameter；Cornea information includes at least: cornea hot spot reflective information；Iris information includes at least: iris edge letter Breath.

Step S304 watches information attentively according to what eye feature information determined target object.

In a kind of optional scheme, server is after receiving eye feature information, and server is according to eyeball tracking Algorithm calculating target object watches information attentively, wherein watching information attentively can be the blinkpunkt of target object on equipment end screen Coordinate information.

Step S306, transmission watch information attentively to equipment end.

It should be noted that equipment end can be communicated through wireless communication with server, wherein wireless communication Mode can include but is not limited to the modes such as WIFI, 3G, 4G, GPRS.In addition it is also necessary to explanation, server will be without that will scheme As information is sent to equipment end, it will only watch information attentively and be sent to equipment end.Server and the traffic rate of equipment end can be with Meet requirement of real-time, and handling capacity is smaller, can satisfy requirement using traffic rate slower GPRS.

Optionally, for server after receiving eye feature information, server calculates target according to eyeball tracking algorithm Object watches information attentively, wherein watching information attentively can be coordinate information of the blinkpunkt of target object on equipment end screen.So Afterwards, the information of watching attentively being calculated is sent to equipment end by server through wireless communication, and letter is watched in equipment end reception attentively Breath.

Based on scheme defined by above-mentioned steps S302 to step S306, it can know, using equipment end and server phase In conjunction with mode, in the image information of equipment end acquisition target object, and the eye for extracting from image information target object is special Reference breath, is then sent to server for eye feature information, is handled by server according to eye feature information, obtain mesh Mark object watches information attentively, and the information of watching attentively of target object is sent to equipment end, and then can be in equipment end displaying target pair Elephant watches information attentively.

It is easily noted that, the acquisition of the image information of target object, the extraction of eye feature information are transported in equipment end Row, and the treatment process of eye feature information is realized in server.Watched attentively due to being handled according to eye feature information The resource of the blinkpunkt algorithm consumption of information is more, therefore, it is handled in server, effectively reduces the money of equipment end Source consumption.In addition, resource consumption is caused if image, which is directly uploaded to server, will occupy more bandwidth, therefore, In the application, equipment end can extract eye feature information from image information, and eye feature information is sent to server, Rather than image information is sent directly to server, it can reduce and image information is directly uploaded to the occupied band of server Width further decreases the resource consumption of equipment end.

In a kind of optional scheme, equipment end sends the eye feature information of target object according to frame format, wherein according to Information is watched attentively according to what eye feature information determined target object, may include steps of:

Step S3020 parses the data of frame format, obtains parsing result, wherein the data of frame format include eye Portion's characteristic information；

Step S3022, verifies parsing result, obtains check results；

Step S3024 determines target object according to eye feature information in the case where check results meet preset condition Watch information attentively.

It should be noted that above-mentioned frame format can be frame format shown in Fig. 2.Optionally, server is receiving eye After portion's characteristic information, eye feature information is parsed, and the check code in frame format is verified.It is logical in check code In the case where crossing verification, server handles the data after parsing, and obtain target object watches information attentively.

Further, after obtaining watching information attentively, server will watch information transmission attentively and watch information attentively to equipment end, specifically Step may include:

Step S3060 determines the corresponding image identification of eye feature information；

Step S3062, acquisition is corresponding with image identification to watch information attentively；

Step S3064, will watch information attentively and image identification format is sent to equipment end.

Optionally, server can send the corresponding eye feature information of current frame image and image according to following frame format Mark: frame head, control word, human eye data and check code, wherein human eye data include watching information attentively.It should be noted that figure As mark can be but be not limited to timestamp, title or be assigned randomly to mark.Wherein, in the feelings that image identification is timestamp Under condition, which is to acquire timestamp corresponding to the time of image information.

Optionally, server believes eye feature based on eyeball tracking algorithm after parsing obtains eye feature information Breath is handled, and obtains watching information attentively, and will watch information attentively according to frame format and feed back to equipment end, at this point, the number in frame format According to watch information attentively.Equipment end is completed to parse to watching information attentively after receiving and watching information attentively, and will watch information biography attentively Pass application end (for example, display equipment).It should be noted that predetermined format includes at least: frame head, control word, human eye data And check code, wherein check code can be but be not limited by CRC (Cyclic Redundancy Check, cyclic redundancy Verification) verify obtained number.After can guaranteeing equipment end and server communication using CRC check, server can be resolved to Correct data, and guarantee the correctness of data, while can also prevent influence of the interference signal to parsing result.

In a kind of optional scheme, server carries out eye feature information after receiving eye feature information Parsing, and the data after parsing are handled, the information of watching attentively of target object is obtained, and information hair will be watched attentively according to frame format It send to equipment end, at this point, the data in frame format are to watch information attentively.Equipment end is completed after receiving and watching information attentively to note Visual information is parsed, and will be watched information attentively and be passed to application end (for example, display equipment).

As shown in the above, eyeball tracking technology can be effectively reduced to the meter of equipment end in scheme provided herein The requirement of calculation ability.In addition, completing the extraction to information is watched attentively, Ke Yiti by server real-time update eyeball tracking algorithm Rise the eyeball tracking processing speed of equipment end.Finally, scheme provided herein is in equipment end operation image Processing Algorithm, Server runs blinkpunkt algorithm, can substantially reduce the handling capacity of communication data, improves traffic rate.

Embodiment 4

According to embodiments of the present invention, a kind of eyeball tracking method embodiment applied to terminal device is provided, wherein Fig. 7 It is the flow chart of the eyeball tracking method according to an embodiment of the present invention applied to terminal device, as shown in fig. 7, this method includes Following steps:

Step S702 extracts the eye feature information of target object from the image information of target object.

It should be noted that the eye feature information of target object includes at least one following: pupil information, cornea letter Breath, eyelid information, eyes image information, watches depth information attentively at iris information, wherein pupil information includes at least: pupil center Position, pupil diameter；Cornea hot spot includes at least: cornea hot spot reflective information；Iris information includes at least: iris edge letter Breath.

In a kind of optional scheme, equipment end has image acquisition device, and image acquisition device can acquire the figure of target object As information.After collecting the image information of target object, equipment end realizes eyeball tracking skill using pupil-corneal reflection method Art, to extract the eye feature information of target object.Optionally, it by taking pupil-corneal reflection method as an example, is adopted in image capture device After the image information for collecting target object, equipment end is believed using the eye feature that image processing algorithm extracts target object Breath, wherein eye feature Informational Expression is the coordinate data in image information, for example, the coordinate of pupil center location, corneal light The coordinate etc. of the center of spot.

Step S704 sends eye feature information to server, wherein server handle to eye feature information Information is watched attentively to target object.

It should be noted that equipment end can be communicated through wireless communication with server, wherein wireless communication Mode can include but is not limited to the modes such as WIFI, 3G, 4G, GPRS.In addition it is also necessary to explanation, in equipment end to service During device sends eye feature information, equipment end is not necessarily to collected whole image information being sent to server, only will Pupil information, cornea information, iris information and eyelid information, eyes image information watch the eye features such as depth information letter attentively Breath is sent to server.When it sends, it can believe according to the eye feature of 1 byte frame head, 1 byte control word, 3 to 6 bytes Breath, 2 byte CRC checks are sent to server, wherein the quantity total amount of transmission is about 10 bytes.In addition, in 90 frames of processing per second In the case where image data, the data volume that equipment end is sent to server is 90*10*8=7.2kbps.It can be seen that equipment end It can satisfy requirement of real-time with the traffic rate of server, and handling capacity is smaller, even if slower using traffic rate GPRS also can satisfy requirement.

In a kind of optional scheme, server is after receiving eye feature information, and server is according to eyeball tracking Algorithm calculating target object watches information attentively, wherein watching information attentively can be the blinkpunkt of target object on equipment end screen Coordinate information.Then, the information of watching attentively being calculated is sent to equipment end, equipment by server through wireless communication End, which receives, watches information attentively.

It should be noted that in above process, server is not necessarily to image information being sent to equipment end, letter will be only watched attentively Breath is sent to equipment end.Server and the traffic rate of equipment end can satisfy requirement of real-time, and handling capacity is smaller, It can satisfy requirement using traffic rate slower GPRS.

Based on scheme defined by above-mentioned steps S702 to step S704, it can know, using equipment end and server phase In conjunction with mode, in the image information of equipment end acquisition target object, and the eye for extracting from image information target object is special Reference breath, is then sent to server for eye feature information, is handled by server according to eye feature information, obtain mesh Mark object watches information attentively, and the information of watching attentively of target object is sent to equipment end, and then can be in equipment end displaying target pair Elephant watches information attentively.

It is easily noted that, the acquisition of the image information of target object, the extraction of eye feature information are transported in equipment end Row, and the treatment process of eye feature information is realized in server.Watched attentively due to being handled according to eye feature information The resource of the blinkpunkt algorithm consumption of information is more, therefore, it is handled in server, effectively reduces the money of equipment end Source consumption.

Embodiment 5

According to embodiments of the present invention, a kind of eyeball tracking method embodiment applied to terminal device is provided, wherein Fig. 8 It is the flow chart of the eyeball tracking method according to an embodiment of the present invention applied to terminal device, as shown in figure 8, this method includes Following steps:

Step S802 obtains the image information that equipment end is sent.

It should be noted that equipment end has image acquisition device, which can acquire the image letter of target object Breath, wherein include the eyes image of target object in the image information of target object.Optionally, in the feelings that equipment end is mobile phone Under condition, image acquisition device can be the camera of mobile phone.Before the resolution ratio of the rear camera of usual mobile phone is higher than mobile phone The resolution ratio of camera is set, but the resolution ratio of the front camera of mobile phone is millions pixel, acquired image information is full The demand of sufficient eyeball tracking technology.In the case where mobile phone has front camera, image acquisition device is selected in the present embodiment The front camera of mobile phone.

In addition it is also necessary to explanation, since the processing capacity of server is stronger, can run more complicated algorithm, Therefore, after collecting the image information of target object, equipment end by the image information of target object be sent to server into Row processing, to achieve the purpose that reduce equipment end resource consumption.

Step S804 extracts eye feature information from image information, forms eye feature information aggregate, wherein image It include eye feature information in information.

In a kind of optional scheme, after receiving image information, server is based on data extraction algorithm and extracts figure As at least one of information in information: pupil information, cornea information, iris information and/or eyelid information, eyes image letter It ceases, watch depth information attentively, and at least one general information is as eye feature information.

Step S806 determines visual information according to eye feature information aggregate.

In a kind of optional scheme, after obtaining eye feature information, server is calculated according to eyeball tracking algorithm Target object watches information attentively, wherein watches information attentively including at least direction of gaze information and blinkpunkt information, wherein watch attentively Point information can be coordinate information of the blinkpunkt of target object on equipment end screen.Then, server will be calculated Watch information attentively and be sent to equipment end through wireless communication, information is watched in equipment end reception attentively, i.e. execution step S808.

It should be noted that since computing capability of the eyeball tracking algorithm to equipment end is more demanding, eyeball tracking algorithm The requirement of kernel and dominant frequency to equipment end is stringent, and the equipment end of middle-end level is difficult to meet the hardware index.Moreover, if Run eyeball tracking algorithm always in equipment end, then it is larger to the resource consumption of equipment end, even result in fever, the card of equipment end ?.And scheme provided herein runs eyeball tracking algorithm in server, it is possible to prevente effectively from the above problem.

Step S808, transmission watch information attentively to equipment end.

It should be noted that equipment end can be communicated through wireless communication with server, wherein wireless communication Mode can include but is not limited to the modes such as WIFI, 3G, 4G, GPRS.

In addition it is also necessary to explanation, server is not necessarily to image information being sent to equipment end, will only watch information transmission attentively To equipment end.Server and the traffic rate of equipment end can satisfy requirement of real-time, and handling capacity is smaller, even if making It also can satisfy requirement with the slower GPRS of traffic rate.

Based on scheme defined by above-mentioned steps S802 to step S808, it can know, using equipment end and server phase In conjunction with mode, get equipment end transmission target object image information after, server is extracted from image information The eye feature information of target object, and determine according to eye feature information the information of watching attentively of target object, it then will watch letter attentively Breath is sent to equipment end.

It is easily noted that, disappears due to being handled to obtain the blinkpunkt algorithm for watching information attentively according to eye feature information The resource of consumption is more, therefore, it is handled in server, effectively reduces the resource consumption of equipment end.In addition, due to clothes Be engaged in device processing capacity it is stronger, more complicated algorithm can be run, therefore, server can directly to target object image letter Breath is handled to obtain eye feature information, and according to eye feature information extraction to information is watched attentively, will finally watch information hair attentively It send to equipment end, further decreases the resource consumption of equipment end.

Embodiment 6

According to embodiments of the present invention, a kind of eyeball tracking Installation practice applied to terminal device, the device are provided It is applied to the eyeball tracking method of terminal device provided by executable embodiment 1.Wherein, Fig. 4 is according to embodiments of the present invention The eyeball tracking device applied to terminal device structural schematic diagram, as shown in figure 4, the device includes: the first acquisition module 401, the first sending module 403 and the first receiving module 405.

Wherein, the first acquisition module 401, for acquiring image information；First sending module 403, for sending eye spy Information aggregate is levied to processor；First receiving module 405, for receive and parse through processor feedback watch information attentively.

It should be noted that above-mentioned first acquisition module 401, the first sending module 403 and the first receiving module 405 are right Should be in the step S102 to step S104 in embodiment 1, example and application scenarios that three modules and corresponding step are realized It is identical, but it is not limited to the above embodiments 1 disclosure of that.

In a kind of optional scheme, the first sending module includes: the first acquisition module, the first determining module and second Sending module.Wherein, first module is obtained, sends information for obtaining；First determining module, for determining that sending information corresponds to Image identification, wherein image identification be used for identify send information transmission sequentially；Second sending module, for letter will to be sent Breath is sent to cloud processor with the first regular fashion.

It should be noted that above-mentioned first obtains module, the first determining module and the second sending module corresponding to implementation Step S1060 to step S1064 in example 1, three modules are identical as example and application scenarios that corresponding step is realized, but It is not limited to the above embodiments 1 disclosure of that.

In a kind of optional scheme, the first determining module includes: the second determining module and the first generation module.Its In, the second determining module, for determining the corresponding timestamp of first frame image for sending information；First generation module is used for root The corresponding image identification of first frame image is generated according to timestamp.Wherein, sending information includes: the first frame head, the first control word, the One check code and image information.

In a kind of optional scheme, the first receiving module include: the second receiving module, parsing module, correction verification module with And third determining module.Wherein, the second receiving module, for receiving the reception information sent in a manner of Second Rule；Parse mould Block receives information for parsing, and receives information and includes the second frame head, the second control word and the second check code and watch letter attentively Breath；Correction verification module verifies the second check code for completing；Third determining module, if the verification for the second check code When as a result meeting preset condition, information is watched attentively by watch that information is determined as target object attentively.

Embodiment 7

According to embodiments of the present invention, a kind of eyeball tracking Installation practice applied to terminal device, the device are provided It is applied to the eyeball tracking method of terminal device provided by executable embodiment 2.Wherein, Fig. 5 is according to embodiments of the present invention The eyeball tracking device applied to terminal device structural schematic diagram, as shown in figure 5, the device includes: the second acquisition module 501, the second extraction module 503, the second sending module 505 and the second receiving module 507.

Wherein, the second acquisition module 501, for acquiring image information；Second extraction module 503 is used for from image information Middle extraction eye feature information forms eye feature information aggregate, wherein includes eye feature information in image information；Second Sending module 505, for sending eye feature information aggregate to server；Second receiving module 507, for receiving and parsing through clothes Business device feedback watches information attentively.

It should be noted that above-mentioned second acquisition module 501, the second extraction module 503, the second sending module 505 and Second receiving module 507 corresponds to the step S902 to step S908 in embodiment 2, and four modules are realized with corresponding step Example it is identical with application scenarios, but be not limited to the above embodiments 2 disclosure of that.

In a kind of optional scheme, the second sending module includes: to obtain module, determining module and sending module.Its In, module is obtained, sends information for obtaining；Determining module sends the corresponding image identification of information for determining, obtains image Identify corresponding eye feature information aggregate, wherein image identification is used to identify the transmission sequence for sending information；Sending module is used Server is sent in information will be sent.Wherein, send information further include: the first frame head, the first control word, the first check code and Eye feature information aggregate.

It should be noted that above-mentioned acquisition module, determining module and sending module correspond to the step in embodiment 2 S2060 to step S2064, three modules are identical as example and application scenarios that corresponding step is realized, but are not limited to above-mentioned 2 disclosure of that of embodiment.

Embodiment 8

According to embodiments of the present invention, a kind of eyeball tracking system embodiment applied to terminal device is provided, wherein figure 6 be the structural schematic diagram of the eyeball tracking system according to an embodiment of the present invention applied to terminal device, as shown in fig. 6, this is System includes: equipment end 601 and server 603.Wherein, equipment end 601, for acquiring the image information of target object, and it is right The eye feature information of target object completes sequence, forms the eye feature information of series, then believes the eye feature of series Breath is sent to server, and receive and parse through server feedback watches information attentively；Server 603 is sent for receiving device end Eye feature information, and eye feature information is handled, obtain target object watches information attentively.

From the foregoing, it will be observed that using equipment end in such a way that server combines, in the image letter of equipment end acquisition target object It ceases, and extracts the eye feature information of target object from image information, eye feature information is then sent to server, by Server is handled according to eye feature information, obtains the information of watching attentively of target object, and watch target object attentively information It is sent to equipment end, and then information can be watched attentively in equipment end displaying target object.

It is easily noted that, the acquisition of the image information of target object, the extraction of eye feature information are transported in equipment end Row, and the treatment process of eye feature information is realized in server.Watched attentively due to being handled according to eye feature information The resource of the blinkpunkt algorithm consumption of information is more, therefore, it is handled in server, effectively reduces the money of equipment end Source consumption.In addition, in this application, equipment end extracts eye feature information from image information, and by eye feature information It is sent to server, rather than image information is sent directly to server, can reduce and image information is directly uploaded to clothes The business occupied bandwidth of device, further decreases the resource consumption of equipment end.

It should be noted that the eye for being applied to terminal device provided in embodiment 1,2 and 4 can be performed in equipment end 601 Ball method for tracing；The eyeball tracking method for being applied to terminal device provided in embodiment 3 and 5 can be performed in server 603.Tool Hold in vivo and be described in detail in embodiment 1-5, details are not described herein.

Embodiment 9

According to another aspect of an embodiment of the present invention, a kind of storage medium is additionally provided, which includes storage Program, wherein program executes the eyeball tracking method for being applied to terminal device provided by embodiment 1 to 5.

Embodiment 10

According to another aspect of an embodiment of the present invention, a kind of processor is additionally provided, which is used to run program, In, the eyeball tracking method for being applied to terminal device provided by embodiment 1 to 5 is executed when program is run.

The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.

In the above embodiment of the invention, it all emphasizes particularly on different fields to the description of each embodiment, does not have in some embodiment The part of detailed description, reference can be made to the related descriptions of other embodiments.

In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, such as the division of the unit, Ke Yiwei A kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or components can combine or Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual Between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of unit or module It connects, can be electrical or other forms.

The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple On unit.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.

It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list Member both can take the form of hardware realization, can also realize in the form of software functional units.

If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words It embodies, which is stored in a storage medium, including some instructions are used so that a computer Equipment (can for personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole or Part steps.And storage medium above-mentioned includes: that USB flash disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic or disk etc. be various to can store program code Medium.

The above is only optional embodiment of the invention, it is noted that for the ordinary skill of the art For personnel, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications It should be regarded as protection scope of the present invention.

Claims

1. a kind of eyeball tracking method applied to terminal device characterized by comprising

Acquire image information；

Described image information is sent to cloud processor；

Receive and parse through the cloud processor feedback watches information attentively.

2. eyeball tracking method according to claim 1, which is characterized in that send described image information to cloud and handle Device, comprising:

It obtains and sends information；

Determine the corresponding image identification of the transmission information, wherein described image mark is for identifying the hair for sending information Send sequence；

The transmission information is sent to the cloud processor with the first regular fashion.

3. eyeball tracking method according to claim 2, which is characterized in that determine the corresponding image mark of the transmission information Know, comprising:

Determine the corresponding timestamp of first frame image for sending information；

The corresponding image identification of the first frame image is generated according to the timestamp.

4. eyeball tracking method according to claim 3, the transmission information includes: the first frame head, the first control word, One check code and described image information.

5. the method according to claim 1, wherein receive and parse through cloud processor feedback watches letter attentively Breath, comprising:

Receive the reception information sent in a manner of Second Rule；

The reception information is parsed, the reception information includes the second frame head, the second control word and the second check code and institute It states and watches information attentively；

Completion verifies second check code；

If the check results of second check code meet preset condition, by the note watched information attentively and be determined as target object Visual information.

6. a kind of eyeball tracking method applied to terminal device characterized by comprising

Obtain the image information that equipment end is sent；

Eye feature information is extracted from described image information, forms eye feature information aggregate, wherein in described image information Include the eye feature information；

Information is watched attentively according to eye feature information aggregate determination；

Information is watched described in transmission attentively to the equipment end.

7. a kind of eyeball tracking method applied to terminal device characterized by comprising

Acquire image information；

The eye feature information aggregate is sent to the server；

Receive and parse through the server feedback watches information attentively.

8. eyeball tracking method according to claim 7, which is characterized in that before forming eye feature information aggregate, Further include:

Corresponding eye feature information is converted by described image information；

By the corresponding eye feature asset data label；

The corresponding eye feature information of configured data label is completed into sequence according to specified sequence.

9. eyeball tracking method according to claim 8, which is characterized in that the corresponding eye feature information includes mesh It marks pupil information, cornea information, facula information, iris information and/or the eyelid information of object, eyes image information, watch depth attentively Spend one or more of information.

10. eyeball tracking method according to claim 7, which is characterized in that the eye feature information aggregate is configured For one or more groups of eye feature information with data label.

11. eyeball tracking method according to claim 7, which is characterized in that send the eye feature information aggregate extremely Server, comprising:

It obtains and sends information；

It determines the corresponding image identification of the transmission information, obtains the corresponding eye feature information aggregate of described image mark, Wherein, described image mark is for identifying the transmission sequence for sending information；

The transmission information is sent to server with the first regular fashion.

12. eyeball tracking method according to claim 11, which is characterized in that determine the corresponding image of the transmission information Mark, comprising:

13. eyeball tracking method according to claim 12, the transmission information include: the first frame head, the first control word, First check code and the eye feature information aggregate.

14. the method according to the description of claim 7 is characterized in that receive and parse through the server feedback watches information attentively, It specifically includes:

Receive the reception information sent in a manner of Second Rule；

Completion verifies second check code；

15. a kind of eyeball tracking device applied to terminal device characterized by comprising

First acquisition module, for acquiring image information；

First sending module, for sending described image information to cloud processor；

First receiving module watches information attentively for receive and parse through cloud processor feedback.

16. a kind of eyeball tracking device applied to terminal device characterized by comprising

Second acquisition module, for acquiring image information；

Second extraction module forms eye feature information aggregate for the extraction eye feature information from described image information, In, it include the eye feature information in described image information；

Second sending module, for sending the eye feature information aggregate to server；

Second receiving module watches information attentively for receive and parse through the server feedback.

17. a kind of eyeball tracking system applied to terminal device characterized by comprising

Equipment end for acquiring image information, and extracts eye feature information from described image information, forms eye feature letter Breath set, is then sent to server for the eye feature information, receive and parse through the server feedback watches information attentively；

The server, the eye feature information sent for receiving the equipment end, and the eye feature information is carried out Processing, obtain target object watches information attentively.

18. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein described program right of execution Benefit require any one of 1 to 14 described in be applied to terminal device eyeball tracking method.

19. a kind of processor, which is characterized in that the processor is for running program, wherein right of execution when described program is run Benefit require any one of 1 to 14 described in be applied to terminal device eyeball tracking method.