WO2021143228A1 - 数据推送方法、装置、电子设备、计算机存储介质和计算机程序 - Google Patents

数据推送方法、装置、电子设备、计算机存储介质和计算机程序 Download PDF

Info

Publication number
WO2021143228A1
WO2021143228A1 PCT/CN2020/121249 CN2020121249W WO2021143228A1 WO 2021143228 A1 WO2021143228 A1 WO 2021143228A1 CN 2020121249 W CN2020121249 W CN 2020121249W WO 2021143228 A1 WO2021143228 A1 WO 2021143228A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
image
display device
face
information display
Prior art date
Application number
PCT/CN2020/121249
Other languages
English (en)
French (fr)
Inventor
赵馨逸
于萌
Original Assignee
北京市商汤科技开发有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京市商汤科技开发有限公司 filed Critical 北京市商汤科技开发有限公司
Priority to JP2021524015A priority Critical patent/JP2022520512A/ja
Publication of WO2021143228A1 publication Critical patent/WO2021143228A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
    • H04N21/44213—Monitoring of end-user related data
    • H04N21/44218—Detecting physical presence or behaviour of the user, e.g. using sensors to detect if the user is leaving the room or changes his face expression during a TV program
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0254—Targeted advertisements based on statistics
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/251—Learning process for intelligent management, e.g. learning user preferences for recommending movies

Definitions

  • the present disclosure relates to computer vision technology, and in particular to a data push method, device, electronic equipment, computer storage medium, and computer program.
  • the embodiment of the present disclosure provides a data push method, and the method includes:
  • the recommended information is determined, and the recommended information is displayed through the information display device.
  • the determining an alternative delivery target according to the stay time includes:
  • the stay duration is greater than or equal to the set duration, and/or the person corresponding to the face within the set range is used as the candidate placement target.
  • the shooting range of the image acquisition device includes an area in front of the information display device
  • the image acquisition device is deployed on the information display device, or the distance between the image acquisition device and the information display device is less than or equal to a set distance threshold.
  • the determining recommendation information according to the candidate placement target includes:
  • At least one candidate placement target in the candidate placement targets is determined to be a final placement target, and the final placement target
  • the distance from the information display device is less than or equal to the set distance
  • the recommendation information is determined for the final delivery target.
  • the determining recommendation information according to the candidate placement target includes:
  • each candidate drop target in the target person image among the candidate drop targets is determined to be the final drop target, and the final drop target is
  • the pixel size corresponding to the face in the target person image is greater than or equal to the set size, or the final placement target is the area corresponding to the face in the target person image, and the reference area in the target person image
  • the proportion of is greater than or equal to the set proportion
  • the recommendation information is determined for the final delivery target.
  • the method further includes:
  • the set size is determined according to the image capture resolution of the image capture device, and the set size is positively correlated with the image capture resolution of the image capture device.
  • the determining, based on the multiple frames of person images, the length of stay of the face in the multiple frames of person images in front of the information display device includes:
  • the length of stay of the human faces in the multiple frames of human images in front of the information display device is determined.
  • the method further includes:
  • a frame of image with the highest face quality is selected.
  • the determining the recommendation information according to the candidate placement target includes: determining the recommendation information based on the face image of the candidate placement target corresponding to the frame of image with the highest face quality.
  • the embodiment of the present disclosure also proposes a data push device, the device includes:
  • the acquisition module is used to acquire multiple frames of human images collected by the image acquisition device;
  • a determining module configured to determine, based on the multiple frames of person images, the length of time that a face in the multiple frames of person images stays in front of the information display device;
  • the processing module is configured to determine the candidate placement target according to the stay time; determine the recommended information according to the candidate placement target, and display the recommended information through the information display device.
  • the processing module is configured to determine an alternative placement target according to the stay time, including:
  • the stay duration is greater than or equal to the set duration, and/or the person corresponding to the face within the set range is used as the candidate placement target.
  • the shooting range of the image acquisition device includes an area in front of the information display device
  • the image acquisition device is deployed on the information display device, or the distance between the image acquisition device and the information display device is less than or equal to a set distance threshold.
  • the processing module is configured to determine recommended information according to the candidate placement target, including:
  • At least one candidate placement target in the candidate placement targets is determined to be a final placement target, and the final placement target
  • the distance from the information display device is less than or equal to the set distance
  • the recommendation information is determined for the final delivery target.
  • the processing module is configured to determine recommended information according to the candidate placement target, including:
  • each candidate drop target in the target person image among the candidate drop targets is determined to be the final drop target, and the final drop target is
  • the pixel size corresponding to the face in the target person image is greater than or equal to the set size, or the final placement target is the area corresponding to the face in the target person image, and the reference area in the target person image
  • the proportion of is greater than or equal to the set proportion
  • the recommendation information is determined for the final delivery target.
  • the processing module is further configured to determine the set size according to the image capture resolution of the image capture device, and the set size is positively correlated with the image capture resolution of the image capture device.
  • the determining module is configured to determine, based on the multiple frames of person images, the length of time that a face in the multiple frames of person images stays in front of the information display device, including:
  • the length of stay of the human faces in the multiple frames of human images in front of the information display device is determined.
  • the determining module is further used for:
  • a frame of image with the highest face quality is selected.
  • the processing module is further configured to: determine the recommendation information according to the face image of the candidate placement target corresponding to the frame of image with the highest face quality.
  • the embodiment of the present disclosure also provides an electronic device, including a processor and a memory for storing a computer program that can be executed by the processor; wherein,
  • the processor executes the computer program, it executes any one of the aforementioned data pushing methods.
  • the embodiment of the present disclosure also provides a computer storage medium on which a computer program is stored.
  • the processor implements any one of the aforementioned data pushing methods.
  • the embodiments of the present disclosure also provide a computer program, which when executed by a processor, causes the processor to implement any one of the aforementioned data pushing methods.
  • the way to determine the alternative placement target according to the length of stay of the face in front of the information display device can be More targeted selection of the target of the recommended information, the higher the effectiveness of the recommended information push, reduces the waste of input resources, and further helps to improve the input-output ratio of the recommended information.
  • FIG. 1 is a flowchart of a data push method according to an embodiment of the disclosure
  • FIG. 2 is a schematic diagram of an application scenario of an embodiment of the disclosure
  • FIG. 3 is a schematic diagram of another application scenario of an embodiment of the disclosure.
  • FIG. 4 is a schematic structural diagram of a data pushing device according to an embodiment of the disclosure.
  • FIG. 5 is a schematic structural diagram of an electronic device according to an embodiment of the disclosure.
  • the terms "including”, “including” or any other variations thereof are intended to cover non-exclusive inclusion, so that a method or device including a series of elements not only includes the explicitly stated Elements, but also include other elements not explicitly listed, or elements inherent to the implementation of the method or device. Without more restrictions, the element defined by the sentence "including a" does not exclude the existence of other related elements (such as steps or steps in the method) in the method or device that includes the element.
  • the unit in the device for example, the unit in the device may be a part of a circuit, a part of a processor, a part of a program or software, etc.).
  • the term "and/or" used in the present disclosure is only used to describe the association relationship of the associated objects, indicating that there can be three relationships, for example, A and/or B, which can mean: A alone exists, and A and B exist at the same time , There are three cases of B alone.
  • the term "at least one" used in the present disclosure means any one of a plurality of elements or any combination of at least two of the plurality of elements, for example, including at least one of A, B, and C, which may Indicates that any one or more elements selected from the set consisting of A, B, and C are included.
  • the data push method provided by the embodiment of the present disclosure includes a series of steps, but the data push method provided by the embodiment of the present disclosure is not limited to the recorded steps.
  • the data push device provided by the embodiment of the present disclosure includes a series of steps.
  • a series of modules, but the data push device provided by the embodiments of the present disclosure is not limited to include the explicitly recorded modules, and may also include modules that need to be set for obtaining relevant information or processing based on the information.
  • Electronic devices such as terminals and servers can be described in the general context of computer system executable instructions (such as program modules) executed by a computer system.
  • program modules may include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types.
  • the computer system/server can be implemented in a distributed cloud computing environment. In the distributed cloud computing environment, tasks are executed by remote processing equipment linked through a communication network.
  • program modules may be located on a storage medium of a local or remote computing system including a storage device.
  • a data push method is proposed.
  • the embodiments of the present disclosure can be applied to scenarios that require recommendation information placement, such as advertisement placement and offline retail.
  • FIG. 1 is a flowchart of a data pushing method according to an embodiment of the present disclosure. As shown in FIG. 1, the data pushing method may include:
  • Step 101 Acquire multiple frames of person images collected by an image collection device.
  • the image capture device refers to a device used to capture video streams or images.
  • the image capture device may be a video detection device, or a capture machine or other device that captures images.
  • the image acquisition device may be arranged around the information display device.
  • the shooting range of the image acquisition device may include the area in front of the information display device.
  • the area in front of the information display device is: the area where the content presented by the information display device can be viewed, for example,
  • the front area of the information display device may be a front area and/or a side front area of the information display device, which is not limited in the embodiment of the present disclosure.
  • the information display device is a device that displays recommended information; for example, the information display device may be a display screen or other types of display devices.
  • the form in which the information display device displays the recommended information may include, but is not limited to, display through pictures, texts, lists, etc.
  • the recommended information can be displayed in a static or dynamic manner.
  • the way of dynamic display includes, but is not limited to, periodic or non-periodical switching of recommended information, for example, switching of recommended information through scrolling, page turning, prominent display, etc.
  • the distance between the image capture device and the information display device is less than or equal to the set distance threshold; it can be understood that when the distance between the image capture device and the information display device is small, the image capture device The shooting range is likely to include the location area where the content presented by the information display device is viewed.
  • the embodiments of the present disclosure facilitate the image collection device to collect images of people in the front area of the information display device.
  • the set distance threshold is positively correlated with the shooting range of the image capture device.
  • the set distance threshold is larger.
  • the shooting range of the image acquisition device may also include the area in front of the information display device; in the case of a small shooting range of the image acquisition device, the distance between the image acquisition device and the information display device is relatively short. It is small, which is conducive to the image capture device shooting the area in front of the information display device.
  • multiple frames of person images can be selected based on face recognition and/or human body recognition among all the image frames of the video stream; you can also select all the images of the video stream first. Part of the image frames are selected from the frames, and then, in the above-mentioned partial image frames, based on face recognition and/or human body recognition, multiple frames of person images are selected.
  • the image capture device is a device that captures images, such as a capture machine, multiple captures can be used to obtain multiple frames of human images.
  • Step 102 Based on the multiple frames of person images, determine the length of stay of the face in the multiple frames of person images in front of the information display device.
  • face detection can be performed on each frame of person images, and on the basis of face detection on each frame of person images, face tracking of multiple frames of person images can be realized; in the embodiments of the present disclosure, there is nothing wrong.
  • the implementation of face detection and face tracking is limited.
  • the length of stay of the human faces in the multiple frames of human images in front of the information display device may be determined.
  • the same face in different frames of person images can be determined, and then, according to the collection time interval of different frames of person images, the person in the multiple frames of person images can be determined The length of time the face stays in front of the image capture device.
  • Step 103 Determine an alternative placement target according to the length of stay.
  • candidate placement targets can be selected.
  • the person corresponding to the human face whose staying time is greater than or equal to the set time may be used as the candidate placement target.
  • the set duration may be a default value, or it may be preset according to actual application requirements; for example, in a situation where the information display device is located in a scene where the flow of people is relatively fast, because the characters have limited time to pay attention to the information Therefore, a lower set duration can be set. In this way, when the stay duration is greater than or equal to the lower set duration, it can be considered that the corresponding person has effective attention to the information display device; when the information display device is located at the flow rate of people In the case of a slower scene, since the characters have enough time to pay attention to the information, a higher set duration can be set.
  • the set time can be set to 3 seconds; when the information display device is located in a bus, In the case of a scene where the flow of people is relatively slow, such as in a building, the setting time can be set to 5 seconds.
  • the people corresponding to the faces in the multi-frame person images can be used as candidate placement targets.
  • the setting range may be a default range, or may be preset according to actual application requirements.
  • the upper limit, lower limit, and interval span of the setting range can be flexibly set; for example, the value method of the lower limit of the setting range can be the same as the value method of the above-mentioned setting duration, that is, when the information display device is located in the traffic speed comparison In the case of a fast scene, the lower limit of the setting range is lower; in the case of the information display device in a scene with a slower flow of people, the lower limit of the setting range is higher.
  • the upper limit of the setting range can also be determined according to the flow rate of the actual scene. Illustratively, when the information display device is located in a scene where the flow rate is relatively fast, since the time for the characters to pay attention to the information is limited, a lower value can be set.
  • the upper limit of the setting range In the case that the stay time is greater than or equal to the upper limit of the lower setting range, it can be considered that the character just stays in front of the information display device for a long time but does not pay attention to the presentation content of the information display device; When the display device is located in a scene with a slower flow of people, since the characters have enough time to pay attention to the information, the upper limit of the higher setting range can be set.
  • the upper limit of the setting range can be set to 1 minute; when the information display equipment is located in a bus, a building, etc., the speed of the flow of people In the case of a relatively slow scene, the upper limit of the setting range can be set to 2 minutes.
  • the upper limit and lower limit of the setting range can be determined first according to the content of the foregoing record, and then the interval span of the setting range can be determined; for example, at the lower limit of the setting range
  • the interval span of the setting range is 76 seconds
  • the lower limit of the setting range is 6 seconds and the upper limit of the setting range is 120 seconds
  • the span of the setting range is 114 seconds.
  • configuration items for the placement of the recommendation information can be set in advance, and the configuration items may include the setting duration, the upper limit of the setting range, the lower limit of the setting range, and the interval span of the setting range.
  • Step 104 Determine recommended information according to the candidate placement target, and display the recommended information through the information display device.
  • the recommendation information can be determined by combining the candidate placement target and/or the relevant data of the final placement target mentioned below, so as to provide more targeted recommendation information for the corresponding placement target.
  • the above-mentioned related data may include, but is not limited to, the facial attributes of the delivery target.
  • the face attribute is used to indicate the identity feature of the placement target
  • the identity feature may specifically include one or a combination of multiple items as exemplified below, for example, an identification such as a face identifier that can effectively distinguish different placement targets; gender; At least one of the basic information of the target, such as age, facial expression, etc.; total consumption and/or single consumption.
  • the identity feature may also include unlisted content, which is not limited here.
  • the selection conditions, methods, etc. of the recommended information are not limited.
  • a display instruction can be sent to the information display device to instruct the information display device to display the recommended information; the information display device can display the recommended information after receiving the display instruction.
  • steps 101 to 104 can be implemented by a processor in an electronic device, and the aforementioned processor can be an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), Digital Signal Processing Device (Digital Signal Processing Device, DSPD), Programmable Logic Device (Programmable Logic Device, PLD), Field Programmable Gate Array (Field Programmable Gate Array, FPGA), Central Processing Unit (Central Processing Unit, CPU), At least one of a controller, a microcontroller, and a microprocessor.
  • ASIC Application Specific Integrated Circuit
  • DSP Digital Signal Processor
  • DSPD Digital Signal Processing Device
  • PLD Programmable Logic Device
  • Field Programmable Gate Array Field Programmable Gate Array
  • FPGA Field Programmable Gate Array
  • CPU Central Processing Unit
  • CPU Central Processing Unit
  • the candidate placement target is determined according to the stay time of the face in front of the information display device. In this way, the target of the recommended information can be selected more pertinently, the effectiveness of the recommended information push is higher, and the waste of input resources is reduced, and further, it is beneficial to improve the input-output ratio of the recommended information.
  • the embodiment of the present disclosure can effectively identify the length of stay of each face in front of the image capture device when the image capture device captures multiple people, that is, the embodiment of the present disclosure can be adapted to the recommendation information placement in a multi-person scene.
  • the distance between each alternative delivery target and the information display device can be used to determine the alternative delivery target.
  • At least one alternative placement target is the final placement target, and the distance between the final placement target and the information display device is less than or equal to the set distance;
  • the recommendation information can be determined for the final delivery target.
  • a set distance may be predetermined, and the set distance represents the maximum distance that the user can effectively pay attention to the presentation content of the information display device; for example, the set distance may be 0.5m, 1m, 2m, 3m, 5m, etc. .
  • the set distance may be the distance between other candidate placement targets (that is, candidate placement targets other than the final placement target) and the information display device.
  • the value of the set distance can be automatically adjusted, for example, based on the number of candidate placement targets, the imaging resolution of the image acquisition device used to collect candidate placement targets, and many other factors that may affect the final placement target screening It is determined by one or a combination of multiple items, which is not limited here.
  • At least one candidate placement target whose distance from the information display device is less than or equal to the set distance may be selected, and the selected at least one candidate placement target may be used as the final placement target; According to the final delivery target, other delivery goals can be determined. It can be seen that the distance between the other delivery goals and the information display device is greater than the set distance.
  • the set distance can be determined according to the presentation size of the information displayed by the information display device; the set distance is positively correlated with the presentation size; here, the information displayed by the information display device can be text or image
  • the presentation size of the information displayed by the information display device can be used to characterize the font size of the text
  • the information displayed by the information display device is an image
  • the information display device The presentation size of the displayed information can be used to characterize the size of the image. For example, when the font of the text displayed by the information display device is an oversized font, the distance is set to 5m; when the font of the text displayed by the information display device is a small font, the distance is set to 1m.
  • the presentation size of the information displayed by the information display device is large, the user can effectively watch the information at a longer distance, and when the presentation size of the information displayed by the information display device is small, the user It may be necessary to effectively view the information at a relatively close distance; therefore, the set distance is positively correlated with the aforementioned presentation size, which can make the set distance meet the actual viewing requirements.
  • the candidate placement target is more likely to be a person who is at an effective distance of attention in front of the information display device, and is more likely to stay on the information display device. Pay attention to the recommended information before; therefore, the above method of determining the final target based on the distance between the candidate target and the information display device can select the target of recommended information in a more targeted manner, and the effectiveness of recommended information push is more effective. High, reducing the waste of resources.
  • the candidate placement may be determined based on the face size of each candidate placement target in the target person’s image.
  • At least one candidate drop target in the target is the final drop target, and the pixel size corresponding to the face of the final drop target in the target person image is greater than or equal to the set size, or the final drop target corresponds to the face in the target person image
  • the proportion of the reference area in the target person’s image is greater than or equal to the set proportion;
  • the recommendation information can be determined for the final delivery target.
  • the target person image refers to an image that contains candidate placement targets collected by the image acquisition device.
  • the target person image may include one candidate placement target or multiple candidate placement targets.
  • the implementation of the present disclosure The example does not limit this.
  • the pixel size represents the size of the pixel value occupied by the face in the image; for example, the pixel size may include the length and/or width of the pixel occupied by the face, and accordingly, the set size may include the length value and/or width value of the pixel .
  • the pixels corresponding to the face can be considered The size is greater than or equal to the set size; otherwise, the pixel size corresponding to the face is smaller than the set size.
  • a face recognition algorithm may be used to determine the circumscribed rectangle of the face along the edge of the face, and further, the pixel value of the length and/or width of the circumscribed rectangle of the face may be taken as the occupancy of the face. The length and/or width of the pixel.
  • the set size may be determined according to the set distance and/or the image capture resolution of the image capture device, the set size and the set distance are negatively correlated, and/or the set size is related to the image capture device The image acquisition resolution is positively correlated.
  • the image capture resolution of the image capture device represents the resolution of the captured image.
  • the distance between the candidate placement target and the information display device may be large.
  • the face in the image of the target person The corresponding pixel size is small. Therefore, setting a smaller setting size is more in line with actual needs; when the setting distance is small, the distance between the alternative placement target and the information display device may be smaller.
  • the pixel size corresponding to the face in the target person image is relatively large. Therefore, setting a larger setting size is more in line with actual needs.
  • the pixel size corresponding to the face in the target person’s image is smaller. Therefore, setting a smaller setting size is more consistent with Practical requirements:
  • the image acquisition resolution of the image acquisition device is large, according to the image acquisition characteristics of the image acquisition device, the pixel size corresponding to the face in the target person image is larger, so set a larger set size , Which is more in line with actual needs.
  • the reference area may represent the entire target person image area, and the target person image area may be determined by the shooting range of the image acquisition device.
  • the entire target person image area may include the human body, the ground, buildings, and animals.
  • An image composed of, plants, etc.; the entire target person image area can be represented by the pixel size of the image, for example, the entire target person image area is 3800*2450 (length * width) area, that is, the entire target person image
  • the pixel length value of is 3800
  • the pixel width value of the entire target person image is 2450.
  • the reference area may represent the sum of all face areas in the entire target person image
  • the entire target person image may include one or more face images.
  • the entire target person may be collected. After the image, the face recognition algorithm is used to detect the size of each face area in the entire target person image, and the sum of all face areas in the entire target person image is calculated by summing.
  • the set ratio may be determined according to the set distance, and the set ratio is negatively related to the set distance. For example, when the setting distance is 5m, the setting ratio is 10%; when the setting distance is 1m, the setting ratio is 30%.
  • the distance between the candidate placement target and the information display device may be relatively large.
  • the final placement target is in the image of the target person.
  • the proportion of the area corresponding to the face in the above reference area may be small. Therefore, setting a smaller setting ratio is more in line with actual needs; in the case of a small setting distance, alternative placement targets and information
  • the distance between the display devices may be small.
  • the proportion of the area corresponding to the face of the target person image in the final target person image may be larger in the above-mentioned reference area. Therefore, Setting a larger setting ratio is more in line with actual needs.
  • the candidate placement target is more likely to be a person who is at an effective following distance in front of the information display device, and is more likely to stay in front of the information display device to pay attention to the recommended information; therefore, the above is based on the candidate placement target
  • the size of the face in the target person's image and the way to determine the final target can be more targeted to select the target of the recommended information, the recommended information push is more effective, and the waste of input resources is reduced.
  • At least one of the candidate placement goals can be determined as the final placement goal based on the first factor and the second factor; here ,
  • the first factor indicates the distance between each candidate placement target and the information display device among the candidate placement targets recorded above, and the second factor indicates that each candidate placement target in the candidate placement target recorded above is in the image of the target person Face size in.
  • a candidate placement target that satisfies both the first condition and the second condition can be selected as the final placement target.
  • the first condition represents the relationship between the candidate placement target and the information display device.
  • the distance of is less than or equal to the set distance
  • the second condition indicates that the pixel size corresponding to the face of the candidate drop target in the target person image is greater than or equal to the set size, or the candidate drop target is the face of the target person image
  • the proportion of the corresponding area in the reference area in the target person image is greater than or equal to the set ratio.
  • weights can be introduced to comprehensively consider the first factor and the second factor; that is, the weights of the first factor and the second factor are set respectively, and the weight value of the first factor and the weight value of the second factor can be the same , Can also be different.
  • the information display equipment and image acquisition equipment are located in a relatively closed and narrow scene such as an elevator. All positions in this scene can be considered as a position that pays attention to the presentation content of the information display equipment.
  • a face is facing the information display device in a scene, the person corresponding to the face is more likely to pay attention to the presentation content of the information display device; in this scenario, if the face is facing the information display device, the face corresponds to Is less likely to be concerned about the content presented by the information display device; on the other hand, if the face is facing the information display device, the face size may be larger; if the face is facing the information display device, the face The size may be small; therefore, in this application scenario, a smaller weight value can be set for the first factor, and a larger weight value can be set for the second factor.
  • the information display device and the image acquisition device are located in a relatively spacious scene on both sides of the road, and the information display device and the image acquisition device are located at a higher level.
  • the image captured by the image acquisition device The size of all face images is relatively small.
  • the face quality of each frame of the human images in the multiple frames of human images can also be determined; furthermore, the face quality of each of the multiple frames of human images can be determined; From the image or the image corresponding to the final delivery target, a frame of image with the highest face quality is selected, so as to facilitate face recognition or subsequent processing according to the selected frame of image with the highest face quality.
  • the face quality of each frame of the human image may indicate the sharpness of the human face, the brightness of the image, the angle of the human face, whether it is occluded or not.
  • FIG. 2 is a schematic diagram of an application scenario of an embodiment of the disclosure
  • FIG. 3 is a schematic diagram of another application scenario of an embodiment of the disclosure.
  • the information display device 1 is located in a high place, and the image acquisition device 2 is located in the information Immediately above the display device; based on the content recorded in the foregoing embodiments of the present disclosure, after processing the images collected by the image capture device 2, alternative drop targets can be determined, and further, the final drop target can also be determined; Select the delivery target or the final delivery target, and the recommended information can be displayed by the information display device 1.
  • the recommended information displayed by the information display device 1 is a vehicle image.
  • the face is facing the information display device 1, and the face image collected by the image acquisition device 2 is a frontal face image; referring to Figure 3, the face side faces the information display device 1, and the person collected by the image acquisition device 2
  • the face image is a side face image; based on the aforementioned content, when evaluating the face quality of the image, the face quality can be evaluated according to the angle of the face; in other indicators (such as face sharpness, image brightness, whether it is When indicators such as occlusion are all equal, the quality of the front face image collected by the image acquisition device 2 in the scene of FIG. 2 is higher than the quality of the side face image collected by the image acquisition device 2 in the scene of FIG. 3.
  • the recommendation information can be determined according to the candidate placement target or the face image of the final placement target corresponding to the frame of image with the highest face quality.
  • an embodiment of the present disclosure proposes a data pushing device.
  • FIG. 4 is a schematic structural diagram of a data pushing device according to an embodiment of the disclosure.
  • the data pushing device may include: an acquisition module 401, a determination module 402, and a processing module 403, where:
  • the obtaining module 401 is used to obtain multiple frames of person images collected by an image collecting device;
  • the determining module 402 is configured to determine, based on the multiple frames of person images, the length of time that a face in the multiple frames of person images stays in front of the information display device;
  • the processing module 403 is configured to determine the candidate placement target according to the stay time; determine the recommended information according to the candidate placement target, and display the recommended information through the information display device.
  • the processing module 403 is configured to determine an alternative placement target according to the stay time, including:
  • the stay duration is greater than or equal to the set duration, and/or the person corresponding to the face within the set range is used as the candidate placement target.
  • the shooting range of the image acquisition device includes an area in front of the information display device
  • the image acquisition device is deployed on the information display device, or the distance between the image acquisition device and the information display device is less than or equal to a set distance threshold.
  • the processing module 403 is configured to determine recommended information according to the candidate placement target, including:
  • At least one candidate placement target in the candidate placement targets is determined to be a final placement target, and the final placement target
  • the distance from the information display device is less than or equal to the set distance
  • the recommendation information is determined for the final delivery target.
  • the processing module 403 is configured to determine recommended information according to the candidate placement target, including:
  • each candidate drop target in the target person image among the candidate drop targets is determined to be the final drop target, and the final drop target is
  • the pixel size corresponding to the face in the target person image is greater than or equal to the set size, or the final placement target is the area corresponding to the face in the target person image, and the reference area in the target person image
  • the proportion of is greater than or equal to the set proportion
  • the recommendation information is determined for the final delivery target.
  • the processing module 403 is further configured to determine the set size according to the image capture resolution of the image capture device, and the set size is positively correlated with the image capture resolution of the image capture device .
  • the determining module 402 is configured to determine, based on the multiple frames of person images, the length of stay of the face in the multiple frames of person images in front of the information display device, including:
  • the acquisition module 401, the determination module 402, and the processing module 403 can all be implemented by a processor in an electronic device.
  • the aforementioned processors can be ASIC, DSP, DSPD, PLD, FPGA, CPU, controller, microcontroller, At least one of the microprocessors.
  • the functional modules in this embodiment may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be realized in the form of hardware or software function module.
  • the integrated unit is implemented in the form of a software function module and is not sold or used as an independent product, it can be stored in a computer readable storage medium.
  • this embodiment is essentially or All or part of this embodiment can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) Or a processor (processor) executes all or part of the steps of the method described in this embodiment.
  • the aforementioned storage media include: U disk, mobile hard disk, read only memory (Read Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program codes.
  • the computer program instructions corresponding to a data push method in this embodiment can be stored on storage media such as optical disks, hard disks, and USB flash drives.
  • the computer program instructions corresponding to a data push method in the storage medium In the case of being read or executed by an electronic device, any one of the data pushing methods of the foregoing embodiments is implemented.
  • FIG. 5 shows an electronic device 500 provided by an embodiment of the present disclosure, which may include: a memory 501 and a processor 502; wherein,
  • the memory 501 is used to store computer programs and data
  • the processor 502 is configured to execute a computer program stored in the memory to implement any data push method in the foregoing embodiments.
  • the aforementioned memory 501 may be a volatile memory (volatile memory), such as RAM; or a non-volatile memory (non-volatile memory), such as ROM, flash memory, or hard disk (Hard Disk). Drive, HDD) or Solid-State Drive (SSD); or a combination of the foregoing types of memories, and provide instructions and data to the processor 502.
  • volatile memory volatile memory
  • non-volatile memory non-volatile memory
  • ROM read-only memory
  • flash memory read-only memory
  • HDD hard disk
  • SSD Solid-State Drive
  • the embodiments of the present disclosure also provide a computer program, which, when executed by a processor, implements any one of the data pushing methods in the foregoing embodiments.
  • the functions or modules contained in the device provided in the embodiments of the present disclosure can be used to execute the methods described in the above method embodiments.
  • the functions or modules contained in the device provided in the embodiments of the present disclosure can be used to execute the methods described in the above method embodiments.
  • the embodiments of the present disclosure can essentially be embodied in the form of a software product.
  • the computer software product is stored in a storage medium (such as ROM/RAM, magnetic disk, and optical disk), and includes several instructions to make a
  • the terminal which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) executes the methods described in the various embodiments of the present disclosure.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Multimedia (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Finance (AREA)
  • Social Psychology (AREA)
  • Game Theory and Decision Science (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computing Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • User Interface Of Digital Computer (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本公开实施例提出了一种数据推送方法、装置、电子设备、计算机存储介质和计算机程序，该方法包括：获取图像采集设备采集的多帧人物图像；基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长；根据所述停留时长，确定备选投放目标；根据所述备选投放目标，确定推荐信息，并通过所述信息展示设备展示所述推荐信息。

Description

数据推送方法、装置、电子设备、计算机存储介质和计算机程序
相关申请的交叉引用
本公开要求于2020年1月17日提交的、申请号为202010054747.0、发明名称为“数据推送方法、装置、电子设备和计算机存储介质”的中国专利申请的优先权，该中国专利申请公开的全部内容以引用的方式并入本文中。
技术领域
本公开涉及计算机视觉技术，尤其涉及一种数据推送方法、装置、电子设备、计算机存储介质和计算机程序。
背景技术
在新零售的行业背景下，线下商业场景中的推荐信息如广告的投放，越来越普及。以各大商场的广告投放为例，目前，大多推荐信息包括商场的促销内容、合作伙伴的宣传内容等，但由于受众群体的多样化，导致该推荐信息往往得不到理想的投放效果，造成投放资源的浪费。
发明内容
本公开实施例提供了一种数据推送方法，所述方法包括：
获取图像采集设备采集的多帧人物图像；
基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长；
根据所述停留时长，确定备选投放目标；
根据所述备选投放目标，确定推荐信息，并通过所述信息展示设备展示所述推荐信息。
可选地，所述根据所述停留时长，确定备选投放目标，包括：
将所述停留时长大于或等于设定时长，和/或，所述停留时长属于设定范围内的人脸对应的人物，作为备选投放目标。
可选地，所述图像采集设备的拍摄范围包括所述信息展示设备的前方区域；
所述图像采集设备部署在所述信息展示设备上，或者，所述图像采集设备与所述信息展示设备之间的距离小于或等于设定距离阈值。
可选地，所述根据所述备选投放目标，确定推荐信息，包括：
根据所述备选投放目标中每个备选投放目标与所述信息展示设备之间的距离，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标与所述信息展示设备之间的距离，小于或等于设定距离；
针对所述最终投放目标，确定所述推荐信息。
可选地，所述根据所述备选投放目标，确定推荐信息，包括：
根据所述备选投放目标中每个备选投放目标在目标人物图像中的人脸尺寸，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标在所述目标人物图像中的人脸对应的像素尺寸大于或等于设定尺寸，或者，所述最终投放目标在所述目标人物图像中的人脸对应的区域，在所述目标人物图像中参考区域的占比，大于或等于设定比例；
针对所述最终投放目标，确定所述推荐信息。
可选地，所述方法还包括：
根据所述图像采集设备的图像采集分辨率，确定所述设定尺寸，所述设定尺寸与所述图像采集设备的图像采集分辨率成正相关。
可选地，所述基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长，包括：
通过对所述多帧人物图像进行人脸跟踪，确定所述多帧人物图像中人脸在所述信息展示设备前的停留时长。
可选地，所述方法还包括：
确定所述多帧人物图像中的每帧人物图像的人脸质量；
在所述备选投放目标对应的图像中，选取出人脸质量最高的一帧图像。
可选地，所述根据所述备选投放目标，确定推荐信息，包括：根据所述人脸质量最高的一帧图像对应的备选投放目标的人脸图像，确定推荐信息。
本公开实施例还提出了一种数据推送装置，所述装置包括：
获取模块，用于获取图像采集设备采集的多帧人物图像；
确定模块，用于基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长；
处理模块，用于根据所述停留时长，确定备选投放目标；根据所述备选投放目标，确定推荐信息，并通过所述信息展示设备展示所述推荐信息。
可选地，所述处理模块用于根据所述停留时长，确定备选投放目标，包括：
将所述停留时长大于或等于设定时长，和/或，所述停留时长属于设定范围内的人脸对应的人物，作为备选投放目标。
可选地，所述图像采集设备的拍摄范围包括所述信息展示设备的前方区域；
所述图像采集设备部署在所述信息展示设备上，或者，所述图像采集设备与所述信息展示设备之间的距离小于或等于设定距离阈值。
可选地，所述处理模块，用于根据所述备选投放目标，确定推荐信息，包括：
根据所述备选投放目标中每个备选投放目标与所述信息展示设备之间的距离，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标与所述信息展示设备之间的距离，小于或等于设定距离；
针对所述最终投放目标，确定所述推荐信息。
可选地，所述处理模块，用于根据所述备选投放目标，确定推荐信息，包括：
根据所述备选投放目标中每个备选投放目标在目标人物图像中的人脸尺寸，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标在所述目标人物图像中的人脸对应的像素尺寸大于或等于设定尺寸，或者，所述最终投放目标在所述目标人物图像中的人脸对应的区域，在所述目标人物图像中参考区域的占比，大于或等于设定比例；
针对所述最终投放目标，确定所述推荐信息。
可选地，所述处理模块，还用于根据所述图像采集设备的图像采集分辨率，确定所述设定尺寸，所述设定尺寸与所述图像采集设备的图像采集分辨率成正相关。
可选地，所述确定模块，用于基于所述多帧人物图像，确定所述多帧人物图像中人 脸在信息展示设备前的停留时长，包括：
通过对所述多帧人物图像进行人脸跟踪，确定所述多帧人物图像中人脸在所述信息展示设备前的停留时长。
可选地，所述确定模块还用于：
确定所述多帧人物图像中的每帧人物图像的人脸质量；
在所述备选投放目标对应的图像中，选取出人脸质量最高的一帧图像。
可选地，所述处理模块还用于：根据所述人脸质量最高的一帧图像对应的备选投放目标的人脸图像，确定推荐信息。
本公开实施例还提供了一种电子设备，包括处理器和用于存储能够由该处理器执行的计算机程序的存储器；其中，
所述处理器在执行所述计算机程序时，执行上述任意一种所述的数据推送方法。
本公开实施例还提供了一种计算机存储介质，其上存储有计算机程序，该计算机程序被处理器执行时，使该处理器实现上述任意一种所述的数据推送方法。
本公开实施例还提供了一种计算机程序，该计算机程序被处理器执行时，使该处理器实现上述任意一种所述的数据推送方法。
在本公开实施例中，由于在信息展示设备前的停留时长是判断用户是否关注推荐信息的重要参考因素，因而，根据人脸在信息展示设备前的停留时长确定备选投放目标的方式，可以更有针对性地选取推荐信息的投放目标，推荐信息推送的有效性更高，减少了投放资源的浪费，进一步地，有利于提升推荐信息的投入产出比。
应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，而非限制本公开。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分，这些附图示出了符合本公开的实施例，并与说明书一起用于说明本公开实施例。
图1为本公开实施例的数据推送方法的流程图；
图2为本公开实施例的一个应用场景的示意图；
图3为本公开实施例的另一个应用场景的示意图；
图4为本公开实施例的数据推送装置的结构示意图；
图5为本公开实施例的电子设备的结构示意图。
具体实施方式
以下结合附图及实施例，对本公开进行进一步详细说明。应当理解，此处所提供的实施例仅仅用以解释本公开，并不用于限制本公开。另外，以下所提供的实施例是用于实施本公开的部分实施例，而非提供实施本公开的全部实施例。在不冲突的情况下，本公开实施例可以任意组合的方式实施。
需要说明的是，在本公开实施例中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的方法或者装置不仅包括所明确记载的要素，而且还包括没有明确列出的其他要素，或者是还包括为实施方法或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个......”限定的要素，并不排除在包括该要素的方法或者装置中还存在另外的相关要素(例如方法中的步骤或者装置中的单元，例如装置中的单元可以是部分电路、部分处理器、部分程序或软件等等)。
本公开中使用的术语“和/或”，仅仅是用于描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本公开中使用的术语“至少一种”表示多种元素中的任意一种或多种元素中的至少两种的任意组合，例如，包括A、B、C中的至少一种，可以表示包括从A、B和C构成的集合中选择的任意一个或多个元素。
例如，本公开实施例提供的数据推送方法包含了一系列的步骤，但是本公开实施例提供的数据推送方法不限于所记载的步骤，同样地，本公开实施例提供的数据推送装置包括了一系列模块，但是本公开实施例提供的数据推送装置不限于包括所明确记载的模块，还可以包括为获取相关信息、或基于信息进行处理时所需要设置的模块。
本公开实施例可以应用于终端和服务器等硬件或硬件组成的计算机系统中，并可以与众多其它通用或专用计算系统环境或配置一起操作，或者可通过处理器执行计算机可执行代码的方式实现本公开实施例。这里，终端可以是瘦客户机、厚客户机、手持或膝上设备、基于微处理器的系统、机顶盒、可编程消费电子产品、网络个人电脑、小型计算机系统，等等，服务器可以是服务器计算机系统、小型计算机系统、大型计算机系统 和包括上述任何系统的分布式云计算环境，等等。
终端、服务器等电子设备可以在由计算机系统执行的计算机系统可执行指令(诸如程序模块)的一般语境下描述。通常，程序模块可以包括例程、程序、目标程序、组件、逻辑、数据结构等等，它们执行特定的任务或者实现特定的抽象数据类型。计算机系统/服务器可以在分布式云计算环境中实施，在分布式云计算环境中，任务是由通过通信网络链接的远程处理设备执行的。在分布式云计算环境中，程序模块可以位于包括存储设备的本地或远程计算系统存储介质上。
在线下商业场景中，越来越多的硬件设备开始具有图像采集能力，这些硬件设备有效的触达了购物中心、门店店铺等区域的人群，这为根据不同人物特点推送不同推荐信息的“新零售”提供了可能。然而，摄像头采集的人流场景是复杂的，同一个时间内往往有多个人出现在拍摄到的视频中，这一复杂环境为推荐信息的投放目标的选择增加了难度。
在本公开的一些实施例中，提出了一种数据推送方法，本公开实施例可以应用于广告投放、线下零售等需要进行推荐信息投放的场景。
图1为本公开实施例的数据推送方法的流程图，如图1所示，该数据推送方法可以包括：
步骤101：获取图像采集设备采集的多帧人物图像。
在本公开实施例中，图像采集设备表示用于采集视频流或图像的设备，例如，图像采集设备可以是视频检测设备，也可以是抓拍机等采集图像的设备。
图像采集设备可以设置于信息展示设备的周围，图像采集设备的拍摄范围可以包括信息展示设备的前方区域，信息展示设备的前方区域为：可以观看到信息展示设备呈现内容的区域，示例性地，信息展示设备的前方区域可以是信息展示设备的正前方区域和/或侧前方区域，本公开实施例并不对此进行限定。
本公开实施例中，信息展示设备为展示推荐信息的设备；示例性地，信息展示设备可以是显示屏或其它类型的显示装置。信息展示设备展示推荐信息的形式，可以包括但不限于通过图文、列表等方式进行展示。在实际展示过程中，可采用静态或是动态展示的方式展示推荐信息。其中，动态展示的方式，包括但不限于推荐信息周期性或非周期性切换，比如，通过滚动、翻页、突出展示等形式进行推荐信息切换。
在一实施方式中，图像采集设备部署在所述信息展示设备上，如此，图像采集设备 的拍摄范围会包括观看信息展示设备的呈现内容的位置区域，结合上述信息展示设备的前方区域的说明，本公开实施例有利于图像采集设备采集到信息展示设备的前方区域的人物图像。
在另一实施方式中，图像采集设备与信息展示设备之间的距离小于或等于设定距离阈值；可以理解的是，在图像采集设备与信息展示设备之间的距离较小时，图像采集设备的拍摄范围很可能会包括观看信息展示设备的呈现内容的位置区域，结合上述信息展示设备的前方区域的说明，本公开实施例有利于图像采集设备采集到信息展示设备的前方区域的人物图像。
这里，设定距离阈值与图像采集设备的拍摄范围成正相关，在图像采集设备的拍摄范围较大的情况下，设定距离阈值较大，这样，即使在图像采集设备与信息展示设备之间的距离较大的情况下，图像采集设备的拍摄范围也可以包括所述信息展示设备的前方区域；在图像采集设备的拍摄范围较小的情况下，图像采集设备与信息展示设备之间的距离较小，有利于图像采集设备拍摄到信息展示设备的前方区域。
这里，在图像采集设备采集到视频流的情况下，可以在视频流的全部图像帧中，基于人脸识别和/或人体识别，选取出多帧人物图像；也可以先在视频流的全部图像帧中选取出部分图像帧，然后，在上述部分图像帧中，基于人脸识别和/或人体识别，选取出多帧人物图像。在另一些实施例中，图像采集设备为抓拍机等采集图像的设备的情况下，可以通过多次抓拍，得到多帧人物图像。
步骤102：基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长。
在实际应用中，可以对每帧人物图像进行人脸检测，在对每帧人物图像进行人脸检测的基础上，可以实现对多帧人物图像的人脸跟踪；本公开实施例中，并不对人脸检测和人脸跟踪的实现方式进行限定。
在一实施方式中，可以通过对多帧人物图像进行人脸跟踪，确定所述多帧人物图像中人脸在所述信息展示设备前的停留时长。具体地说，通过对多帧人物图像进行人脸跟踪，可以确定出在不同帧人物图像的同一人脸，然后，根据采集不同帧人物图像的采集时间间隔，可以确定出多帧人物图像中人脸在图像采集设备前的停留时长。
步骤103：根据所述停留时长，确定备选投放目标。
具体地说，可以根据多帧人物图像中人脸在信息展示设备前的停留时长，在多帧人 物图像中人脸对应的人物中，选取出备选投放目标。
对于本步骤的实现方式，在一个示例中，可以在多帧人物图像中人脸对应的人物中，将停留时长大于或等于设定时长的人脸对应的人物，作为备选投放目标。
本公开实施例中，设定时长可以是默认值，或者，可以根据实际应用需求预先设置；示例性地，在信息展示设备位于人流速度比较快的场景的情况下，由于人物关注信息的时间有限，因而，可以设置较低的设定时长，如此，在停留时长大于或等于较低的设定时长的情况下，可以认为相应的人物对信息展示设备产生有效关注；在信息展示设备位于人流速度比较慢的场景的情况下，由于人物关注信息的时间充足，因而，可以设置较高的设定时长，如此，在停留时长大于或等于较高的设定时长的情况下，可以认为相应的人物对信息展示设备产生有效关注；在具体的示例中，在信息展示设备位于地铁站点等人流速度比较快的场景的情况下，设定时长可以设置为3秒；在信息展示设备位于公交车内、建筑物内等人流速度比较慢的场景的情况下，设定时长可以设置为5秒。
可以理解的是，在信息展示设备前的停留时长较短的用户关注推荐信息的可能性较低，在信息展示设备前的停留时长较长的用户更有可能关注推荐信息，因而，在人脸在信息展示设备前的停留时长大于或等于设定时长的情况下，将该人脸对应的人物作为推荐信息的备选投放目标，可以更有针对性地选取推荐信息的投放目标。
对于本步骤的实现方式，在另一个示例中，可以在多帧人物图像中人脸对应的人物中，将停留时长属于设定范围内的人脸对应的人物，作为备选投放目标。
本公开实施例中，设定范围可以是默认范围，或者，可以根据实际应用需求预先设置。
设定范围的上限、下限和区间跨度均可以灵活设置；示例性地，设定范围的下限的取值方式可以与上述设定时长的取值方式相同，即，在信息展示设备位于人流速度比较快的场景的情况下，设定范围的下限较低；在信息展示设备位于人流速度比较慢的场景的情况下，设定范围的下限较高。
设定范围的上限也可以根据实际场景的人流速度进行确定，示例性地，在信息展示设备位于人流速度比较快的场景的情况下，由于人物关注信息的时间有限，因而，可以设置较低的设定范围的上限，在停留时长大于或等于较低的设定范围的上限的情况下，可以认为人物仅仅是长时间停留在信息展示设备前但并未关注信息展示设备的呈现内容；在信息展示设备位于人流速度比较慢的场景的情况下，由于人物关注信息的时间充 足，因而，可以设置较高的设定范围的上限，此时，在停留时长大于或等于较高的设定范围的上限的情况下，可以认为人物仅仅是长时间停留在信息展示设备前但并未关注信息展示设备的呈现内容。在具体的示例中，在信息展示设备位于地铁站点等人流速度比较快的场景的情况下，设定范围的上限可以设置为1分钟；在信息展示设备位于公交车内、建筑物内等人流速度比较慢的场景的情况下，设定范围的上限可以设置为2分钟。
对于设定范围的区间跨度的确定方式，示例性地，可以根据前述记载的内容先确定设定范围的上限和下限，进而可以确定出设定范围的区间跨度；例如，在设定范围的下限为4秒，且设定范围的上限为80秒的情况下，设定范围的区间跨度为76秒；在设定范围的下限为6秒，且设定范围的上限为120秒的情况下，设定范围的区间跨度为114秒。
根据前述记载的内容，可以看出，在信息展示设备前的停留时长属于设定范围内的用户关注推荐信息的可能性较高，在信息展示设备前的停留时长不属于设定范围内的用户关注推荐信息的可能性较低，因而，在人脸在信息展示设备前的停留时长属于设定范围的情况下，将该人脸对应的人物作为推荐信息的备选投放目标，可以更有针对性地选取推荐信息的投放目标。
在实际应用中，在推荐信息投放前，可以预先设置推荐信息的投放的配置项，该配置项可以包括设定时长、设定范围的上限、设定范围的下限和设定范围的区间跨度。
步骤104：根据所述备选投放目标，确定推荐信息，并通过所述信息展示设备展示所述推荐信息。
推荐信息可以结合备选投放目标和/或下文提到的最终投放目标的相关数据来确定，从而更有针对性地为相应的投放目标提供推荐信息。在一种实现方式中，上述相关数据可以包括但不限于投放目标的人脸属性。其中，人脸属性用于表示投放目标的身份特征，该身份特征具体可以包括如下例举的一项或是多项的组合，比如，诸如人脸标识等能够有效区分不同投放目标的标识；性别、年龄、表情等投放目标的基础信息中的至少一项；消费总额和/或消费单数。当然该身份特征还可以包括未例举的内容，在此不予限定。在本公开实施例中，对于推荐信息的选取条件、方式等不予限定。
在实际应用中，在确定推荐信息后，可以向信息展示设备发送显示指令，以指示信息展示设备展示推荐信息；信息展示设备在接收到显示指令后，可以展示上述推荐信息。
在实际应用中，步骤101至步骤104可以利用电子设备中的处理器实现，上述处理 器可以为特定用途集成电路(Application Specific Integrated Circuit，ASIC)、数字信号处理器(Digital Signal Processor，DSP)、数字信号处理装置(Digital Signal Processing Device，DSPD)、可编程逻辑装置(Programmable Logic Device，PLD)、现场可编程门阵列(Field Programmable Gate Array，FPGA)、中央处理器(Central Processing Unit，CPU)、控制器、微控制器、微处理器中的至少一种。
可以看出，在本公开实施例中，由于在信息展示设备前的停留时长是判断用户是否关注推荐信息的重要参考因素，因而，根据人脸在信息展示设备前的停留时长确定备选投放目标的方式，可以更有针对性地选取推荐信息的投放目标，推荐信息推送的有效性更高，减少了投放资源的浪费，进一步地，有利于提升推荐信息的投入产出比。
另外，本公开实施例可以在图像采集设备拍摄到多人的情况下，有效识别每个人脸在图像采集设备前的停留时长，即，本公开实施例能够适应于多人场景的推荐信息投放。
对于根据备选投放目标，确定推荐信息的实现方式，在第一个示例中，可以根据备选投放目标中每个备选投放目标与信息展示设备之间的距离，确定备选投放目标中的至少一个备选投放目标为最终投放目标，最终投放目标与信息展示设备之间的距离，小于或等于设定距离；
在得到最终投放目标后，可以针对最终投放目标，确定所述推荐信息。
在一实施方式中，可以预先确定设定距离，该设定距离表示用户对信息展示设备的呈现内容形成有效关注的最大距离；例如设定距离可以是0.5m、1m、2m、3m、5m等。在一种可能的实现方式中，设定距离可以为其他备选投放目标(即除最终投放目标以外的备选投放目标)与信息展示设备之间的距离。在实际应用中，设定距离的取值可以自动调整，比如，基于备选投放目标的数量、用于采集备选投放目标的图像采集设备的成像分辨率等诸多可能影响最终投放目标筛选因素中的一项或是多项的结合来确定，在此不予限定。
然后，可以在各备选投放目标中，选取与信息展示设备之间的距离小于或等于设定距离的至少一个备选投放目标，并将选取出的至少一个备选投放目标作为最终投放目标；根据最终投放目标便可以确定出其他投放目标，可以看出，其他投放目标与信息展示设备之间的距离大于设定距离。
在实际应用中，可以根据信息展示设备展示的信息的呈现尺寸，确定所述设定距离；所述设定距离与所述呈现尺寸成正相关；这里，信息展示设备展示的信息可以为 文字、图像等信息，在信息展示设备展示的信息为文字的情况下，信息展示设备展示的信息的呈现尺寸可以用于表征文字的字体大小；在信息展示设备展示的信息为图像的情况下，信息展示设备展示的信息的呈现尺寸可以用于表征图像的大小。例如，在信息展示设备展示的文字的字体为超大号字体的情况下，设定距离为5m；在信息展示设备展示的文字的字体为小号字体的情况下，设定距离为1m。
可以理解的是，在信息展示设备展示的信息的呈现尺寸较大的情况下，用户可以在较远的距离有效地观看信息，在信息展示设备展示的信息的呈现尺寸较小的情况下，用户可以需要在较近的距离有效地观看信息；因而，设定距离与上述呈现尺寸成正相关，可以使得设定距离符合实际观看需求。
可以看出，在备选投放目标与信息展示设备之间的距离比较小的情况下，备选投放目标更可能是在信息展示设备前处于有效关注距离的人，更有可能停留在信息展示设备前对推荐信息进行关注；因而，上述根据备选投放目标与信息展示设备之间的距离确定最终投放目标的方式，可以更有针对性地选取推荐信息的投放目标，推荐信息推送的有效性更高，减少了投放资源的浪费。
对于根据备选投放目标，确定推荐信息的实现方式，在第二个示例中，可以根据所述备选投放目标中每个备选投放目标在目标人物图像中的人脸尺寸，确定备选投放目标中的至少一个备选投放目标为最终投放目标，最终投放目标在目标人物图像中的人脸对应的像素尺寸大于或等于设定尺寸，或者，最终投放目标在目标人物图像中的人脸对应的区域，在目标人物图像中参考区域的占比，大于或等于设定比例；
在得到最终投放目标后，可以针对最终投放目标，确定所述推荐信息。
本公开实施例中，目标人物图像表示图像采集设备采集到的包含备选投放目标的图像，在目标人物图像中可以包括一个备选投放目标，也可以包括多个备选投放目标，本公开实施例并不对此进行限定。像素尺寸表征人脸在图像中占据像素值大小；示例性地，像素尺寸可以包括人脸占据的像素的长度和/或宽度，相应地，设定尺寸可以包括像素的长度值和/或宽度值。如果人脸占据的像素的长度大于或等于设定尺寸中的长度值，和/或，如果人脸占据的像素的宽度大于或等于设定尺寸中的宽度值，则可以认为人脸对应的像素尺寸大于或等于设定尺寸；否则，人脸对应的像素尺寸小于设定尺寸。
在具体实现的实施例中，可以采用人脸识别算法，沿人脸边缘确定出人脸外切矩形，进而，可以将人脸外切矩形的长和/或宽的像素值作为人脸占据的像素的长度和/ 或宽度。
在一实施方式中，可以根据设定距离和/或图像采集设备的图像采集分辨率，确定设定尺寸，设定尺寸与设定距离成负相关，和/或，设定尺寸与图像采集设备的图像采集分辨率成正相关。
这里，图像采集设备的图像采集分辨率表示采集到的图像的分辨率，图像采集设备的图像采集分辨率越高，则图像采集设备采集的图像的像素长度值和/或像素宽度值越大。
可以理解的是，在设定距离较大的情况下，备选投放目标与信息展示设备之间的距离可能较大，此时，根据图像采集设备的图像采集特点，目标人物图像中的人脸对应的像素尺寸较小，因而，设置较小的设定尺寸，更加符合实际需求；在设定距离较小的情况下，备选投放目标与信息展示设备之间的距离可能较小，此时，根据图像采集设备的图像采集特点，目标人物图像中的人脸对应的像素尺寸较大，因而，设置较大的设定尺寸，更加符合实际需求。
在图像采集设备的图像采集分辨率较小的情况下，根据图像采集设备的图像采集特点，目标人物图像中的人脸对应的像素尺寸较小，因而，设置较小的设定尺寸，更加符合实际需求；在图像采集设备的图像采集分辨率较大的情况下，根据图像采集设备的图像采集特点，目标人物图像中的人脸对应的像素尺寸较大，因而，设置较大的设定尺寸，更加符合实际需求。
在一个示例中，参考区域可以表示整个目标人物图像的区域，目标人物图像的区域可以由图像采集设备的拍摄范围确定，例如，整个目标人物图像的区域可以是包含人体、地面、建筑物、动物、植物等构成的一幅图像；整个目标人物图像的区域可以用图像的像素尺寸进行表示，例如，整个目标人物图像的区域为3800*2450(长*宽)的区域，即，整个目标人物图像的像素长度值为3800，整个目标人物图像的像素宽度值为2450。
在另一个示例中，参考区域可以表示整个目标人物图像中所有人脸区域的总和，整个目标人物图像可以包括一个或多个人脸的图像，在实际的实施例中，可以在采集到整个目标人物图像后，利用人脸识别算法检测出整个目标人物图像中每个人脸区域的大小，通过求和计算得到整个目标人物图像中所有人脸区域的总和。
在一实施方式中，可以根据设定距离确定设定比例，设定比例与设定距离成负 相关。例如，设定距离为5m时，设定比例为10％；设定距离为1m时，设定比例为30％。
可以理解的是，在设定距离较大的情况下，备选投放目标与信息展示设备之间的距离可能较大，此时，根据图像采集设备的图像采集特点，最终投放目标在目标人物图像中的人脸对应的区域在上述参考区域的占比，可能较小，因而，设置较小的设定比例，更加符合实际需求；在设定距离较小的情况下，备选投放目标与信息展示设备之间的距离可能较小，此时，根据图像采集设备的图像采集特点，最终投放目标在目标人物图像中的人脸对应的区域在上述参考区域的占比，可能较大，因而，设置较大的设定比例，更加符合实际需求。
可以看出，在备选投放目标在目标人物图像中的人脸对应的像素尺寸比较大的情况下，或者，在备选投放目标在目标人物图像中的人脸对应的区域在上述参考区域的占比比较大的情况下，备选投放目标更可能是在信息展示设备前处于有效关注距离的人，更有可能停留在信息展示设备前对推荐信息进行关注；因而，上述根据备选投放目标在目标人物图像中的人脸大小，确定最终投放目标的方式，可以更有针对性地选取推荐信息的投放目标，推荐信息推送的有效性更高，减少了投放资源的浪费。
对于根据备选投放目标，确定推荐信息的实现方式，在第三个示例中，可以根据第一因素和第二因素，确定备选投放目标中的至少一个备选投放目标为最终投放目标；这里，第一因素表示上述记载的备选投放目标中每个备选投放目标与信息展示设备之间的距离，第二因素表示上述记载的备选投放目标中每个备选投放目标在目标人物图像中的人脸尺寸。
在一个示例中，可以在备选投放目标中，选取出同时满足第一条件和第二条件的备选投放目标为最终投放目标，这里，第一条件表示备选投放目标与信息展示设备之间的距离小于或等于设定距离，第二条件表示备选投放目标在目标人物图像中的人脸对应的像素尺寸大于或等于设定尺寸，或者，备选投放目标在目标人物图像中的人脸对应的区域，在目标人物图像中参考区域的占比，大于或等于设定比例。
在另一个示例中，可以引入权重对第一因素和第二因素进行综合考虑；即，针对第一因素和第二因素各自设置权重，第一因素的权重值和第二因素的权重值可以相同，也可以不同。在一个应用场景中，信息展示设备和图像采集设备处于电梯等较为封闭狭小的场景中，在这种场景中的所有位置都可以认为是对信息展示设备的呈现内容形成关注的位置，在这种场景中如果人脸正对信息展示设备，则该人脸对应的人物更有可能对信息展示设备的呈现内容形成关注；在这种场景中如果人脸侧对信息展示设备，则该人 脸对应的人物对信息展示设备的呈现内容形成关注的可能性较低；另一方面，如果人脸正对信息展示设备，则人脸尺寸可能较大；如果人脸侧对信息展示设备，则人脸尺寸可能较小；因此，在该应用场景中，可以针对第一因素设置较小的权重值，针对第二因素设置较大的权重值。
在另一个应用场景中，信息展示设备和图像采集设备处于道路两侧等较为宽敞的场景中，且信息展示设备和图像采集设备位于较高处，在这种场景中图像采集设备所拍摄到的所有人脸图像的尺寸均比较小，在这种场景中如果人脸对应的人物与信息展示设备之间的距离较小，则更可能对信息展示设备的呈现内容形成关注；反之，如果人脸对应的人物与信息展示设备之间的距离较大，则有可能看不清楚信息展示设备的呈现内容，即，对信息展示设备的呈现内容形成关注的可能性较低；因此，在该应用场景中，可以针对第一因素设置较大的权重值，针对第二因素设置较小的权重值。
在实际应用中，在对所述多帧人物图像进行人脸跟踪的情况下，还可以确定多帧人物图像中的每帧人物图像的人脸质量；进而，可以在所有备选投放目标对应的图像或最终投放目标对应的图像中，选取出人脸质量最高的一帧图像，以便于根据该选取出的人脸质量最高的一帧图像进行人脸识别或后续处理。这里，每帧人物图像的人脸质量可以表示人脸清晰度、图像亮度、人脸的角度、是否被遮挡等指标。
图2为本公开实施例的一个应用场景的示意图，图3为本公开实施例的另一个应用场景的示意图，图2和图3中，信息展示设备1位于高处，图像采集设备2位于信息展示设备的正上方；基于本公开实施例前述记载的内容，对图像采集设备2采集到的图像进行处理后，可以确定出备选投放目标，进一步地，还可以确定出最终投放目标；针对备选投放目标或最终投放目标，可以通过信息展示设备1展示推荐信息，图2和图3中，信息展示设备1展示的推荐信息为车辆图像。
参照图2，人脸正对信息展示设备1，图像采集设备2采集到的人脸图像为正面人脸图像；参照图3，人脸侧对信息展示设备1，图像采集设备2采集到的人脸图像为侧面人脸图像；基于前述记载的内容，在评估图像的人脸质量时，可以根据人脸的角度来评估人脸质量；在其它指标(如人脸清晰度、图像亮度、是否被遮挡等指标)均相等的情况下，图2的场景中利用图像采集设备2采集的正面人脸图像的质量高于图3的场景中利用图像采集设备2采集的侧面人脸图像的质量。
进一步地，在选取出人脸质量最高的一帧图像后，可以根据该人脸质量最高的一帧图像对应的备选投放目标或最终投放目标的人脸图像，确定推荐信息。
在前述实施例提出的数据推送方法的基础上，本公开实施例提出了一种数据推送装置。
图4为本公开实施例的数据推送装置的结构示意图，如图4所示，该数据推送装置可以包括：获取模块401、确定模块402和处理模块403，其中，
获取模块401，用于获取图像采集设备采集的多帧人物图像；
确定模块402，用于基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长；
处理模块403，用于根据所述停留时长，确定备选投放目标；根据所述备选投放目标，确定推荐信息，并通过所述信息展示设备展示所述推荐信息。
可选地，所述处理模块403用于根据所述停留时长，确定备选投放目标，包括：
将所述停留时长大于或等于设定时长，和/或，所述停留时长属于设定范围内的人脸对应的人物，作为备选投放目标。
可选地，所述图像采集设备的拍摄范围包括所述信息展示设备的前方区域；
所述图像采集设备部署在所述信息展示设备上，或者，所述图像采集设备与所述信息展示设备之间的距离小于或等于设定距离阈值。
可选地，所述处理模块403，用于根据所述备选投放目标，确定推荐信息，包括：
根据所述备选投放目标中每个备选投放目标与所述信息展示设备之间的距离，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标与所述信息展示设备之间的距离，小于或等于设定距离；
针对所述最终投放目标，确定所述推荐信息。
可选地，所述处理模块403，用于根据所述备选投放目标，确定推荐信息，包括：
根据所述备选投放目标中每个备选投放目标在目标人物图像中的人脸尺寸，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标在所述目标人物图像中的人脸对应的像素尺寸大于或等于设定尺寸，或者，所述最终投放目标在所述目标人物图像中的人脸对应的区域，在所述目标人物图像中参考区域的占比，大于或等于设定比例；
针对所述最终投放目标，确定所述推荐信息。
可选地，所述处理模块403，还用于根据所述图像采集设备的图像采集分辨率，确定所述设定尺寸，所述设定尺寸与所述图像采集设备的图像采集分辨率成正相关。
可选地，所述确定模块402，用于基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长，包括：
通过对所述多帧人物图像进行人脸跟踪，确定所述多帧人物图像中人脸在所述信息展示设备前的停留时长。
实际应用中，获取模块401、确定模块402和处理模块403均可以利用电子设备中的处理器实现，上述处理器可以为ASIC、DSP、DSPD、PLD、FPGA、CPU、控制器、微控制器、微处理器中的至少一种。
另外，在本实施例中的各功能模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。
所述集成的单元如果以软件功能模块的形式实现并非作为独立的产品进行销售或使用的情况下，可以存储在一个计算机可读取存储介质中，基于这样的理解，本实施例本质上或者说本实施例的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或processor(处理器)执行本实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
具体来讲，本实施例中的一种数据推送方法对应的计算机程序指令可以被存储在光盘，硬盘，U盘等存储介质上，在存储介质中的与一种数据推送方法对应的计算机程序指令被一电子设备读取或被执行的情况下，实现前述实施例的任意一种数据推送方法。
基于前述实施例，参见图5，其示出了本公开实施例提供的一种电子设备500，可以包括：存储器501和处理器502；其中，
所述存储器501，用于存储计算机程序和数据；
所述处理器502，用于执行所述存储器中存储的计算机程序，以实现前述实施例的任意一种数据推送方法。
在实际应用中，上述存储器501可以是易失性存储器(volatile memory)，例如RAM；或者非易失性存储器(non-volatile memory)，例如ROM，快闪存储器(flash memory)，硬盘(Hard Disk Drive，HDD)或固态硬盘(Solid-State Drive，SSD)；或者上述种类的存储器的组合，并向处理器502提供指令和数据。
上述处理器502可以为ASIC、DSP、DSPD、PLD、FPGA、CPU、控制器、微控制器、微处理器中的至少一种。可以理解的是，对于不同的增强现实云平台，用于实现上述处理器功能的电子器件还可以为其它，本公开实施例不作具体限定。
本公开实施例还提供了一种计算机程序，该计算机程序被处理器执行时，实现前述实施例的任意一种数据推送方法。
在一些实施例中，本公开实施例提供的装置具有的功能或包含的模块可以用于执行上文方法实施例描述的方法，其具体实现可以参照上文方法实施例的描述，为了简洁，这里不再赘述。
上文对各个实施例的描述倾向于强调各个实施例之间的不同之处，其相同或相似之处可以互相参考，为了简洁，本文不再赘述。
本公开所提供的各方法实施例中所揭露的方法，在不冲突的情况下可以任意组合，得到新的方法实施例。
本公开所提供的各产品实施例中所揭露的特征，在不冲突的情况下可以任意组合，得到新的产品实施例。
本公开所提供的各方法或设备实施例中所揭露的特征，在不冲突的情况下可以任意组合，得到新的方法实施例或设备实施例。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述方法实施例可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本公开实施例本质上可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本公开各个实施例所述的方法。
上面结合附图对本公开的实施例进行了描述，但是本公开并不局限于上述的具体实施方式，上述的具体实施方式仅仅是示意性的，而不是限制性的，本领域的普通技术人员在本公开的启示下，在不脱离本公开宗旨和权利要求所保护的范围的情况下，还 可做出很多形式，这些均属于本公开的保护之内。

Claims (21)

  1. 一种数据推送方法，所述方法包括：
    获取图像采集设备采集的多帧人物图像；
    基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长；
    根据所述停留时长，确定备选投放目标；
    根据所述备选投放目标，确定推荐信息，并通过所述信息展示设备展示所述推荐信息。
  2. 根据权利要求1所述的方法，其中，所述根据所述停留时长，确定备选投放目标，包括：
    将所述停留时长大于或等于设定时长，和/或，所述停留时长属于设定范围内的人脸对应的人物，作为备选投放目标。
  3. 根据权利要求1或2所述的方法，其中，所述图像采集设备的拍摄范围包括所述信息展示设备的前方区域；
    所述图像采集设备部署在所述信息展示设备上，或者，所述图像采集设备与所述信息展示设备之间的距离小于或等于设定距离阈值。
  4. 根据权利要求1至3任一项所述的方法，其中，所述根据所述备选投放目标，确定推荐信息，包括：
    根据所述备选投放目标中每个备选投放目标与所述信息展示设备之间的距离，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标与所述信息展示设备之间的距离，小于或等于设定距离；
    针对所述最终投放目标，确定所述推荐信息。
  5. 根据权利要求1至3任一项所述的方法，其中，所述根据所述备选投放目标，确定推荐信息，包括：
    根据所述备选投放目标中每个备选投放目标在目标人物图像中的人脸尺寸，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标在所述目标人物图像中的人脸对应的像素尺寸大于或等于设定尺寸，或者，所述最终投放目标在所述目标人物图像中的人脸对应的区域，在所述目标人物图像中参考区域的占比，大于或等于设定比例；
    针对所述最终投放目标，确定所述推荐信息。
  6. 根据权利要求5所述的方法，所述方法还包括：
    根据所述图像采集设备的图像采集分辨率，确定所述设定尺寸，所述设定尺寸与所述图像采集分辨率成正相关。
  7. 根据权利要求1至6任一项所述的方法，其中，所述基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长，包括：
    通过对所述多帧人物图像进行人脸跟踪，确定所述多帧人物图像中人脸在所述信息展示设备前的停留时长。
  8. 根据权利要求7所述的方法，所述方法还包括：
    确定所述多帧人物图像中的每帧人物图像的人脸质量；
    在所述备选投放目标对应的图像中，选取出人脸质量最高的一帧图像。
  9. 根据权利要求8所述的方法，其中，所述根据所述备选投放目标，确定推荐信息，包括：根据所述人脸质量最高的一帧图像对应的备选投放目标的人脸图像，确定推荐信息。
  10. 一种数据推送装置，所述装置包括：
    获取模块，用于获取图像采集设备采集的多帧人物图像；
    确定模块，用于基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长；
    处理模块，用于根据所述停留时长，确定备选投放目标；根据所述备选投放目标，确定推荐信息，并通过所述信息展示设备展示所述推荐信息。
  11. 根据权利要求10所述的装置，其中，所述处理模块用于根据所述停留时长，确定备选投放目标，包括：
    将所述停留时长大于或等于设定时长，和/或，所述停留时长属于设定范围内的人脸对应的人物，作为备选投放目标。
  12. 根据权利要求10或11所述的装置，其中，所述图像采集设备的拍摄范围包括所述信息展示设备的前方区域；
    所述图像采集设备部署在所述信息展示设备上，或者，所述图像采集设备与所述信息展示设备之间的距离小于或等于设定距离阈值。
  13. 根据权利要求10至12任一项所述的装置，其中，所述处理模块，用于根据所述备选投放目标，确定推荐信息，包括：
    根据所述备选投放目标中每个备选投放目标与所述信息展示设备之间的距离，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标与所述信息展示设备之间的距离，小于或等于设定距离；
    针对所述最终投放目标，确定所述推荐信息。
  14. 根据权利要求10至12任一项所述的装置，其中，所述处理模块，用于根据所述备选投放目标，确定推荐信息，包括：
    根据所述备选投放目标中每个备选投放目标在目标人物图像中的人脸尺寸，确定所述备选投放目标中的至少一个备选投放目标为最终投放目标，所述最终投放目标在所述目标人物图像中的人脸对应的像素尺寸大于或等于设定尺寸，或者，所述最终投放目标在所述目标人物图像中的人脸对应的区域，在所述目标人物图像中参考区域的占比，大于或等于设定比例；
    针对所述最终投放目标，确定所述推荐信息。
  15. 根据权利要求14所述的装置，其中，所述处理模块，还用于根据所述图像采集设备的图像采集分辨率，确定所述设定尺寸，所述设定尺寸与所述图像采集分辨率成正相关。
  16. 根据权利要求10至15任一项所述的装置，其中，所述确定模块，用于基于所述多帧人物图像，确定所述多帧人物图像中人脸在信息展示设备前的停留时长，包括：
    通过对所述多帧人物图像进行人脸跟踪，确定所述多帧人物图像中人脸在所述信息展示设备前的停留时长。
  17. 根据权利要求16所述的装置，其中，所述确定模块还用于：
    确定所述多帧人物图像中的每帧人物图像的人脸质量；
    在所述备选投放目标对应的图像中，选取出人脸质量最高的一帧图像。
  18. 根据权利要求17所述的装置，其中，所述处理模块还用于：根据所述人脸质量最高的一帧图像对应的备选投放目标的人脸图像，确定推荐信息。
  19. 一种电子设备，包括处理器和用于存储能够由该处理器执行的计算机程序的存储器；其中，
    所述处理器在执行所述计算机程序时，执行权利要求1-9任一项所述的数据推送方法。
  20. 一种计算机存储介质，其上存储有计算机程序，其中，该计算机程序被处理器执行时，使该处理器实现权利要求1-9任一项所述的数据推送方法。
  21. 一种计算机程序，该计算机程序被处理器执行时，使该处理器实现权利要求1-9任一项所述的数据推送方法。
PCT/CN2020/121249 2020-01-17 2020-10-15 数据推送方法、装置、电子设备、计算机存储介质和计算机程序 WO2021143228A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2021524015A JP2022520512A (ja) 2020-01-17 2020-10-15 データプッシュ方法、装置、電子デバイス、コンピュータ記憶媒体、及びコンピュータプログラム

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010054747.0 2020-01-17
CN202010054747.0A CN111246296A (zh) 2020-01-17 2020-01-17 数据推送方法、装置、电子设备和计算机存储介质

Publications (1)

Publication Number Publication Date
WO2021143228A1 true WO2021143228A1 (zh) 2021-07-22

Family

ID=70868910

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/121249 WO2021143228A1 (zh) 2020-01-17 2020-10-15 数据推送方法、装置、电子设备、计算机存储介质和计算机程序

Country Status (3)

Country Link
JP (1) JP2022520512A (zh)
CN (1) CN111246296A (zh)
WO (1) WO2021143228A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115146183A (zh) * 2022-08-01 2022-10-04 南京日报发展有限责任公司 一种基于目标路径的信息推荐方法

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111246296A (zh) * 2020-01-17 2020-06-05 北京市商汤科技开发有限公司 数据推送方法、装置、电子设备和计算机存储介质
CN117522476A (zh) * 2020-11-23 2024-02-06 支付宝(杭州)信息技术有限公司 一种对象展示方法、装置及电子设备
CN113065456A (zh) * 2021-03-30 2021-07-02 上海商汤智能科技有限公司 信息提示方法、装置、电子设备和计算机存储介质
CN113076479A (zh) * 2021-04-14 2021-07-06 北京地平线信息技术有限公司 信息推荐方法、装置、信息展示牌设备、介质和电子设备
CN113489895B (zh) * 2021-06-23 2022-05-31 荣耀终端有限公司 确定推荐场景的方法及电子设备

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130198003A1 (en) * 2010-09-30 2013-08-01 Nec Soft, Ltd. Advertisement delivery target identifying apparatus, advertisement delivery apparatus, advertisement delivery target identifying method, advertisement delivery method, program, and recording medium
WO2017080173A1 (zh) * 2015-11-11 2017-05-18 乐视控股（北京）有限公司 基于自然信息识别的推送系统和方法及一种客户端
CN108335138A (zh) * 2018-02-01 2018-07-27 百度在线网络技术（北京）有限公司 信息推送方法和装置
CN110264268A (zh) * 2019-06-19 2019-09-20 京东方科技集团股份有限公司 广告投放装置、方法、设备及其存储介质
CN111246296A (zh) * 2020-01-17 2020-06-05 北京市商汤科技开发有限公司 数据推送方法、装置、电子设备和计算机存储介质

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8224078B2 (en) * 2000-11-06 2012-07-17 Nant Holdings Ip, Llc Image capture and identification system and process
JP2011017883A (ja) * 2009-07-09 2011-01-27 Nec Soft Ltd 訴求対象特定システム、訴求対象特定方法、広告出力システム及び広告出力方法
CN102201187A (zh) * 2010-03-23 2011-09-28 深圳华北工控股份有限公司 一种广告定向投放方法
JP2015228145A (ja) * 2014-06-02 2015-12-17 隆 平林 表示装置、デジタル表示システム及びデジタル表示プログラム
CN104881642B (zh) * 2015-05-22 2018-10-26 海信集团有限公司 一种内容推送方法、装置及设备
CN105825408A (zh) * 2016-05-16 2016-08-03 刘冰 一种广告投放处理方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130198003A1 (en) * 2010-09-30 2013-08-01 Nec Soft, Ltd. Advertisement delivery target identifying apparatus, advertisement delivery apparatus, advertisement delivery target identifying method, advertisement delivery method, program, and recording medium
WO2017080173A1 (zh) * 2015-11-11 2017-05-18 乐视控股（北京）有限公司 基于自然信息识别的推送系统和方法及一种客户端
CN108335138A (zh) * 2018-02-01 2018-07-27 百度在线网络技术（北京）有限公司 信息推送方法和装置
CN110264268A (zh) * 2019-06-19 2019-09-20 京东方科技集团股份有限公司 广告投放装置、方法、设备及其存储介质
CN111246296A (zh) * 2020-01-17 2020-06-05 北京市商汤科技开发有限公司 数据推送方法、装置、电子设备和计算机存储介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115146183A (zh) * 2022-08-01 2022-10-04 南京日报发展有限责任公司 一种基于目标路径的信息推荐方法
CN115146183B (zh) * 2022-08-01 2024-04-26 南京日报发展有限责任公司 一种基于目标路径的信息推荐方法

Also Published As

Publication number Publication date
JP2022520512A (ja) 2022-03-31
CN111246296A (zh) 2020-06-05

Similar Documents

Publication Publication Date Title
WO2021143228A1 (zh) 数据推送方法、装置、电子设备、计算机存储介质和计算机程序
US20230334092A1 (en) Automated media analysis for sponsor valuation
JP4702877B2 (ja) 表示装置
US8290281B2 (en) Selective presentation of images
KR101759453B1 (ko) 자동 이미지 크로핑 및 공유
JP5224360B2 (ja) 電子広告装置、電子広告方法及びプログラム
US6873710B1 (en) Method and apparatus for tuning content of information presented to an audience
CN108965982A (zh) 视频录制方法、装置、电子设备和可读存储介质
US8805123B2 (en) System and method for video recognition based on visual image matching
JP6672305B2 (ja) オブジェクト検出に基づき外挿画像を生成する方法及び装置
CN111861572B (zh) 广告投放方法、装置、电子设备及计算机可读存储介质
CN110264268B (zh) 广告投放装置、方法、设备及其存储介质
KR102097781B1 (ko) 객체 정보 제공 장치 및 방법
JP2011070629A5 (zh)
WO2022095818A1 (en) Methods and systems for crowd motion summarization via tracklet based human localization
CN111246110B (zh) 图像的输出方法和装置、存储介质及电子装置
JP2013195725A (ja) 画像表示システム
JP5962383B2 (ja) 画像表示システムおよび画像処理装置
CN110019951B (zh) 一种生成视频缩略图的方法及设备
CN113947795A (zh) 口罩佩戴检测方法、装置、设备及存储介质
CN113132744A (zh) 直播弹幕的处理方法、模型、电子装置及计算机存储介质
US20150332345A1 (en) Advertisement selection and model augmentation based upon physical characteristics of a viewer
EP2071511A1 (en) Method and device for generating a sequence of images of reduced size
CN116958854A (zh) 视频封面的确定方法、装置、设备、介质和程序产品
CN117834888A (zh) 一种图像编码方法及其电子设备、存储介质

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2021524015

Country of ref document: JP

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20913351

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20913351

Country of ref document: EP

Kind code of ref document: A1