Method, apparatus and computer-readable storage medium for determining pedestrian line-of-sight convergence zone

Abstract

This invention relates to the field of automotive technology and discloses a method, apparatus, and computer-readable storage medium for determining pedestrian gaze clusters. First, pedestrian images of a test area and a corresponding region model are acquired. Then, based on the pedestrian images and the region model, pedestrian gaze intersections within a display area are determined. Finally, the density of these pedestrian gaze intersections within the display area is used to determine the pedestrian gaze cluster. Applying this invention allows for a more scientific determination of pedestrian gaze clusters in surrounding passageways within a display area. This enables the placement of displayed items within these clusters, ensuring they are seen by more pedestrians and attracting their attention.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, specifically to a method, apparatus, and computer-readable storage medium for determining a pedestrian line-of-sight convergence zone. Background Technology

[0002] Shopping malls, as commercial complexes integrating dining, clothing, and entertainment, attract a large and diverse range of users. These highly concentrated user groups make them crucial promotional venues for new energy vehicles. Therefore, most new energy vehicle brands establish experience stores in shopping malls to connect with customers. However, unlike traditional goods, the presence of cars in shopping malls presents challenges due to their large size and limited space for display items. Determining the most easily visible location within the experience store to place the cars, ensuring they are within the sight of more pedestrians and attracting their attention, has become a pressing technical challenge. Summary of the Invention

[0003] In view of the above problems, embodiments of the present invention provide a method, apparatus and computer-readable storage medium for determining pedestrian line-of-sight clusters, to solve the problem in the prior art of how to determine the area in a store that is most easily observed by pedestrians in a shopping mall environment.

[0004] According to one aspect of the present invention, a method for determining a pedestrian visual convergence zone is provided. The method is used to determine the pedestrian visual convergence zone of a region to be measured, the region to be measured including a display area and a surrounding passageway area, comprising:

[0005] Obtain pedestrian images of the region to be tested and the region model corresponding to the region to be tested;

[0006] Based on the pedestrian images and the region model, determine the pedestrian line-of-sight intersection points within the display area;

[0007] Based on the density of the pedestrian line-of-sight intersections within the display area, the pedestrian line-of-sight concentration area within the display area is determined.

[0008] In one implementation, after the step of determining the pedestrian sight-gathering area within the display area based on the density of the sight intersections, the method further includes:

[0009] Obtain the attributes of the items to be displayed in the display area, wherein the attributes of the items to be displayed include at least one of the following: the area occupied by the items to be displayed, the accessible range for pedestrians, or the safe distance between the items to be displayed and surrounding items.

[0010] Based on the attributes of the item to be displayed and the pedestrian eye focus area, a placement area for the item to be displayed is generated.

[0011] In one implementation, the step of determining the intersection point of pedestrian gazes within the display area based on the pedestrian image and the region model specifically includes:

[0012] Based on the pedestrian images, estimate the line of sight for each pedestrian;

[0013] Based on the estimated pedestrian line of sight and the area model, the effective pedestrian line of sight for each pedestrian in the display area is determined;

[0014] Mark the intersection of the effective pedestrian lines of sight of all pedestrians as the pedestrian line of sight intersection point within the display area.

[0015] In one implementation, the pedestrian gaze includes the gaze direction; the step of determining the effective pedestrian gaze of each pedestrian within the display area based on the estimated pedestrian gaze and the region model specifically includes:

[0016] Based on the area model, determine the transparent boundaries within the display area;

[0017] Based on the transparent boundary and the direction of sight, determine the transparent pedestrian sightlines of all pedestrians; the transparent pedestrian sightlines include: the sightline information of the pedestrians' transparent observation display area;

[0018] Based on the visible pedestrian line of sight and the preset line of sight observation threshold, the effective pedestrian line of sight for each pedestrian corresponding to the visible boundary is determined.

[0019] In one implementation, the step of estimating the pedestrian's gaze based on the pedestrian image further includes:

[0020] The pedestrian image is input into a pre-trained gaze recognition model, which outputs the pedestrian's head pose.

[0021] Based on the pedestrian's head posture and the region model, the pedestrian's line of sight is predicted.

[0022] In one implementation, the step of determining the pedestrian gaze concentration area within the display area based on the density of pedestrian gaze intersections within the display area specifically includes:

[0023] If the display area includes a transparent boundary, then the pedestrian gaze concentration area within the display area is determined based on the density of the pedestrian gaze intersection points corresponding to the transparent boundary.

[0024] If the display area includes two or more transparent boundaries, then according to the density of the pedestrian line-of-sight intersection points corresponding to each transparent boundary, the pedestrian line-of-sight gathering area of ​​each transparent boundary is calculated, and the intersection of the pedestrian line-of-sight gathering areas of all transparent boundaries is taken as the final pedestrian line-of-sight gathering area.

[0025] According to another aspect of the present invention, a device for determining a pedestrian line-of-sight convergence zone is provided. The device is used to determine the pedestrian line-of-sight convergence zone of a region to be measured, comprising:

[0026] The first acquisition module is used to acquire pedestrian images of the area to be tested and the area model corresponding to the area to be tested.

[0027] The line-of-sight intersection determination module is used to determine the line-of-sight intersection points of pedestrians within the display area based on the pedestrian image and the region model;

[0028] The pedestrian gaze concentration zone determination module is used to determine the pedestrian gaze concentration zone in the display area based on the density of the pedestrian gaze intersection points in the display area.

[0029] In one implementation, the device for determining the pedestrian line-of-sight convergence zone further includes:

[0030] The second acquisition module is used to acquire the attributes of the items to be displayed in the display area. The attributes of the items to be displayed include at least one of the following: the area occupied by the items to be displayed, the accessible range for pedestrians, or the safe distance between the items to be displayed and surrounding items.

[0031] The placement area generation module is used to generate a placement area for the item to be displayed based on the attributes of the item to be displayed and the pedestrian line of sight gathering area.

[0032] In one implementation, the line-of-sight intersection determination module is further configured to:

[0033] Based on the pedestrian images, estimate the line of sight for each pedestrian;

[0034] Based on the estimated pedestrian line of sight and the area model, the effective pedestrian line of sight for each pedestrian in the display area is determined;

[0035] Mark the intersection of the effective pedestrian lines of sight of all pedestrians as the pedestrian line of sight intersection point within the display area.

[0036] In one implementation, the pedestrian line-of-sight determination module is further used for:

[0037] Obtain the boundary of the display area, which includes a first boundary that is visible to pedestrians and a second boundary that is not visible to pedestrians;

[0038] If the direction of any pedestrian's gaze is pointing towards the second boundary, then the gaze of any pedestrian corresponding to that gaze direction is removed.

[0039] In one implementation, the pedestrian line-of-sight determination module is further used for:

[0040] The pedestrian image is input into a pre-trained gaze recognition model, which outputs the pedestrian's head posture.

[0041] Based on the pedestrian's head posture and the region model, the pedestrian's line of sight is predicted.

[0042] In one implementation, the step of determining the pedestrian gaze concentration area within the display area based on the density of pedestrian gaze intersections within the display area specifically includes:

[0043] If the display area includes a transparent boundary, then the pedestrian gaze concentration area within the display area is determined based on the density of the pedestrian gaze intersection points corresponding to the transparent boundary.

[0044] If the display area includes two or more transparent boundaries, then according to the density of the pedestrian line-of-sight intersection points corresponding to each transparent boundary, the pedestrian line-of-sight gathering area of ​​each transparent boundary is calculated, and the intersection of the pedestrian line-of-sight gathering areas of all transparent boundaries is taken as the final pedestrian line-of-sight gathering area.

[0045] According to another aspect of the present invention, a device for determining a pedestrian line-of-sight convergence zone is provided, comprising:

[0046] The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus.

[0047] The memory is used to store at least one executable instruction, which causes the processor to perform the following steps:

[0048] Obtain pedestrian images of the region to be tested and the region model corresponding to the region to be tested;

[0049] Based on the pedestrian images and the region model, determine the pedestrian line-of-sight intersection points within the display area;

[0050] Based on the density of the pedestrian line-of-sight intersections within the display area, the pedestrian line-of-sight concentration area within the display area is determined.

[0051] In one implementation, after the step of determining the pedestrian sight-gathering area within the display area based on the density of the sight intersections, the method further includes:

[0052] Obtain the attributes of the items to be displayed in the display area, wherein the attributes of the items to be displayed include at least one of the following: the area occupied by the items to be displayed, the accessible range for pedestrians, or the safe distance between the items to be displayed and surrounding items.

[0053] Based on the attributes of the item to be displayed and the pedestrian eye focus area, a placement area for the item to be displayed is generated.

[0054] In one implementation, the step of determining the intersection point of pedestrian gazes within the display area based on the pedestrian image and the region model specifically includes:

[0055] Based on the pedestrian images, estimate the line of sight for each pedestrian;

[0056] Based on the estimated line of sight of each pedestrian and the display area of ​​the area model, determine the effective line of sight of each pedestrian.

[0057] Mark the intersection of the effective pedestrian lines of sight of all pedestrians as the pedestrian line of sight intersection point within the display area.

[0058] In one implementation, the pedestrian gaze includes the gaze direction; the step of determining the effective pedestrian gaze of each pedestrian within the display area based on the estimated pedestrian gaze and the region model specifically includes:

[0059] Based on the area model, determine the transparent boundaries within the display area;

[0060] Based on the transparent boundary and the direction of sight, determine the transparent pedestrian sightlines of all pedestrians; the transparent pedestrian sightlines include: the sightline information of the pedestrians' transparent observation display area;

[0061] Based on the visible pedestrian line of sight and the preset line of sight observation threshold, the effective pedestrian line of sight for each pedestrian corresponding to the visible boundary is determined.

[0062] In one implementation, the step of estimating the pedestrian's gaze based on the pedestrian image further includes:

[0063] The pedestrian image is input into a pre-trained gaze recognition model, which outputs the pedestrian's head pose.

[0064] The pedestrian's line of sight is predicted based on the pedestrian's head posture and the region model.

[0065] In one implementation, the step of determining the pedestrian gaze concentration area within the display area based on the density of pedestrian gaze intersections within the display area specifically includes:

[0066] If the display area includes a transparent boundary, then the pedestrian gaze concentration area within the display area is determined based on the density of the pedestrian gaze intersection points corresponding to the transparent boundary.

[0067] If the display area includes two or more transparent boundaries, then according to the density of the pedestrian line-of-sight intersection points corresponding to each transparent boundary, the pedestrian line-of-sight gathering area of ​​each transparent boundary is calculated, and the intersection of the pedestrian line-of-sight gathering areas of all transparent boundaries is taken as the final pedestrian line-of-sight gathering area.

[0068] According to another aspect of the present invention, a computer-readable storage medium is provided, the storage medium storing at least one executable instruction that causes a device / apparatus for determining a pedestrian line-of-sight convergence zone to perform the following operations:

[0069] Obtain pedestrian images of the region to be tested and the region model corresponding to the region to be tested;

[0070] Based on the pedestrian images and the region model, determine the pedestrian line-of-sight intersection points within the display area;

[0071] Based on the density of the pedestrian line-of-sight intersections within the display area, the pedestrian line-of-sight concentration area within the display area is determined.

[0072] In one implementation, after the step of determining the pedestrian sight-gathering area within the display area based on the density of the sight intersections, the method further includes:

[0073] Obtain the attributes of the items to be displayed in the display area, wherein the attributes of the items to be displayed include at least one of the following: the area occupied by the items to be displayed, the accessible range for pedestrians, or the safe distance between the items to be displayed and surrounding items.

[0074] Based on the attributes of the item to be displayed and the pedestrian eye focus area, a placement area for the item to be displayed is generated.

[0075] In one implementation, the step of determining the intersection point of pedestrian gazes within the display area based on the pedestrian image and the region model specifically includes:

[0076] Based on the pedestrian images, estimate the line of sight for each pedestrian;

[0077] Based on the estimated pedestrian line of sight and the area model, the effective pedestrian line of sight for each pedestrian in the display area is determined;

[0078] Mark the intersection of the effective pedestrian lines of sight of all pedestrians as the pedestrian line of sight intersection point within the display area.

[0079] In one implementation, the pedestrian gaze includes the gaze direction; the step of determining the effective pedestrian gaze of each pedestrian within the display area based on the estimated pedestrian gaze and the region model specifically includes:

[0080] Based on the area model, determine the transparent boundaries within the display area;

[0081] Based on the transparent boundary and the direction of sight, determine the transparent pedestrian sightlines of all pedestrians; the transparent pedestrian sightlines include: the sightline information of the pedestrians' transparent observation display area;

[0082] Based on the visible pedestrian line of sight and the preset line of sight observation threshold, the effective pedestrian line of sight for each pedestrian corresponding to the visible boundary is determined.

[0083] In one implementation, the step of estimating the pedestrian's gaze based on the pedestrian image further includes:

[0084] The pedestrian image is input into a pre-trained gaze recognition model, which outputs the pedestrian's head pose.

[0085] Based on the pedestrian's head posture and the region model, the pedestrian's line of sight is predicted.

[0086] In one implementation, the step of determining the pedestrian gaze concentration area within the display area based on the density of pedestrian gaze intersections within the display area specifically includes:

[0087] If the display area includes a transparent boundary, then the pedestrian gaze concentration area within the display area is determined based on the density of the pedestrian gaze intersection points corresponding to the transparent boundary.

[0088] If the display area includes two or more transparent boundaries, then according to the density of the pedestrian line-of-sight intersection points corresponding to each transparent boundary, the pedestrian line-of-sight gathering area of ​​each transparent boundary is calculated, and the intersection of the pedestrian line-of-sight gathering areas of all transparent boundaries is taken as the final pedestrian line-of-sight gathering area.

[0089] This invention provides a method, apparatus, and computer-readable storage medium for determining pedestrian gaze clusters. The method involves acquiring pedestrian images of a test area and a corresponding region model. Based on the pedestrian images and the region model, pedestrian gaze intersections within a display area are determined. Finally, the density of these intersections determines the pedestrian gaze cluster within the display area. Applying this invention allows for a more scientific determination of pedestrian gaze clusters in surrounding passageways within a display area. This enables the placement of displayed items within these clusters, ensuring they are visible to more pedestrians and attracting their attention.

[0090] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0091] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0092] Figure 1 A flowchart of a first embodiment of a method for determining a pedestrian line-of-sight convergence zone provided by the present invention is shown;

[0093] Figure 2 This diagram illustrates the generation of a pedestrian's line-of-sight intersection point according to the present invention.

[0094] Figure 3 This invention provides another schematic diagram for generating the line-of-sight intersection point of pedestrians.

[0095] Figure 4 A flowchart is shown for another embodiment of a method for determining a pedestrian line-of-sight convergence zone provided by the present invention;

[0096] Figure 5 A flowchart of yet another embodiment of a method for determining a pedestrian line-of-sight convergence zone provided by the present invention is shown;

[0097] Figure 6 A schematic diagram of the structure of a first embodiment of a pedestrian line-of-sight concentration zone determination device provided by the present invention is shown;

[0098] Figure 7 A schematic diagram of an embodiment of a pedestrian sight focus area determination device provided by the present invention is shown. Detailed Implementation

[0099] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0100] Figure 1 The flowchart illustrates a first embodiment of a method for determining a pedestrian visual convergence zone provided by the present invention. This method is executed by a device for determining a pedestrian visual convergence zone. The method is used to determine the pedestrian visual convergence zone of a test area, which includes a display area and surrounding passageways, such as... Figure 1 As shown, the method for determining the pedestrian line-of-sight convergence zone includes the following steps:

[0101] Step 110: Obtain pedestrian images of the area to be tested and the corresponding region model of the area to be tested.

[0102] The pedestrian images include at least images of pedestrians' heads. These images can be captured by installing cameras in the area to be tested. Specifically, the cameras can be a combination of multiple bullet cameras or PTZ cameras to accurately acquire pedestrian images of pedestrians in the area to be tested.

[0103] The region model is established using a planar graphical method, that is, the facility structure within the area to be measured is represented by a planar graphical representation. After obtaining pedestrian images, the pedestrian images are determined at the corresponding positions on the region model.

[0104] Step 120: Based on the pedestrian image and the region model, determine the pedestrian line-of-sight intersection points within the display area.

[0105] Before determining the intersection point of pedestrian lines of sight within the display area, it is necessary to first determine the pedestrian lines of sight within the display area. Then, based on the pedestrian lines of sight, the intersection point of pedestrian lines of sight within the display area is determined. It should be noted that since a single pedestrian line of sight within the display area can only represent the direction and approximate area of ​​the pedestrian's gaze, and cannot precisely pinpoint the focal point of the pedestrian's gaze, in practical applications, multiple pedestrian lines of sight within the display area are used to determine the intersection point of pedestrian lines of sight, thereby estimating the most likely location of the pedestrian's line of sight.

[0106] Specifically, a visual representation method is used to define the pedestrian's line of sight within the display area. This involves representing the pedestrian's line of sight as lines on a planar graphical model of the area. The line of sight includes both the starting point and the direction of the line of sight. The starting point can be represented by a dot, or as shown below. Figure 2 and 3In this method, a human figure diagram is used to represent the starting point of the line of sight, and the direction away from the human figure diagram is the direction of the line of sight.

[0107] The method for determining the pedestrian line-of-sight convergence area provided in this application embodiment reflects the relationship between pedestrians, display areas, and surrounding passage areas by using a line-of-sight graphical representation on a region model.

[0108] It should be noted that in practical applications, after confirming a pedestrian's line of sight, the intersection point of that line of sight with the lines of sight of other pedestrians can be determined. A single line-of-sight intersection point can be generated on the line of sight of a single pedestrian, for example, as shown below. Figure 2 As shown, pedestrian line of sight A intersects with pedestrian line of sight B only once. Therefore, only one line of sight intersection point can be determined on pedestrian line of sight A. Furthermore, multiple line of sight intersection points can also occur on the line of sight of a single pedestrian, for example, as... Figure 3 As shown, pedestrian line of sight A intersects with pedestrian line of sight B, pedestrian line of sight C, and pedestrian line of sight D. Therefore, three line of sight intersections are generated on pedestrian line of sight A.

[0109] The intersection of sight lines does not indicate the actual position that the pedestrian is looking at; rather, it is determined by using a large amount of data statistics to identify the position within the display area that is most easily seen by pedestrians.

[0110] Step 130: Determine the pedestrian sight gathering area in the display area based on the density of the pedestrian sight intersection points in the display area.

[0111] It should be noted that in step 120, since the intersection of sight lines does not necessarily mean that the corresponding pedestrian is looking at that location, in practical applications, it is necessary to determine the pedestrian sight gathering area based on the density of the intersection of sight lines. For example, based on a preset first threshold, when the density of the intersection of sight lines is greater than the first threshold, the corresponding area is determined as the pedestrian sight gathering area.

[0112] It should be noted that the number of pedestrian gaze clusters is not limited to one; there can be multiple clusters. For example, in practical applications, the density of gaze intersections in multiple areas may exceed the first threshold. Alternatively, in practical applications, areas with a gaze intersection density exceeding the second threshold can be identified as candidate areas. Then, based on the attributes of these candidate areas, a certain number of pedestrian gaze clusters can be determined. The attributes of the candidate areas can be based on their area, shape, or location within the display area, or they can be sorted according to the density of gaze intersections within the candidate areas.

[0113] This invention provides a method for determining pedestrian gaze concentration zones. First, it acquires pedestrian images of a test area and a corresponding region model. Then, based on the pedestrian images and the region model, it determines pedestrian gaze intersection points within a display area. Finally, it determines the pedestrian gaze concentration zone within the display area based on the density of these intersection points. Applying this invention allows for a more scientific determination of pedestrian gaze concentration zones in surrounding passageways within a display area, enabling the placement of displayed items within these zones. This ensures that the displayed items are visible to more pedestrians, thereby attracting their attention.

[0114] Figure 4 A flowchart illustrating another embodiment of a method for determining a pedestrian line-of-sight convergence zone provided by the present invention is shown. This method for determining a pedestrian line-of-sight convergence zone is performed by a device for determining a pedestrian line-of-sight convergence zone. Figure 4 As shown, the method for determining the pedestrian line-of-sight convergence zone includes the following steps:

[0115] Step 410: Obtain pedestrian images of the area to be tested and the corresponding region model of the area to be tested.

[0116] Step 420: Based on the pedestrian image and the region model, determine the pedestrian line-of-sight intersection points within the display area.

[0117] Step 430: Determine the pedestrian sight gathering area in the display area based on the density of the pedestrian sight intersection points in the display area.

[0118] Step 440: Obtain the attributes of the items to be displayed in the display area. The attributes of the items to be displayed include at least one of the following: the area occupied by the items to be displayed, the accessible range for pedestrians, and the safe distance between the items to be displayed and surrounding items.

[0119] The pedestrian-accessible range refers to the minimum distance a pedestrian can approach the item to be displayed. For example, in the actual display process, if the item to be displayed cannot be touched by pedestrians, then a prohibited area is generally set up. Of course, in the actual display process, there are also items that pedestrians can directly touch. For example, if the item to be displayed is clothing, pedestrians can pick it up and put it down directly. Or, if the item to be displayed is a car, pedestrians can not only touch it, but under certain conditions, they can even enter the car to experience it. In this case, the pedestrian-accessible range can be set to 0.

[0120] It should be noted that, in the actual display process, the item to be displayed can be a movable or mobile item. For example, the item to be displayed is a robot that can move within a certain range, or a robotic arm that can move. Alternatively, the item to be displayed is a car whose doors can be opened within a certain range. In this case, it is necessary to ensure that there are no other exhibits or store facilities within a certain area around the item to be displayed.

[0121] Step 450: Based on the attributes of the item to be displayed and the pedestrian eye focus area, generate the placement area for the item to be displayed.

[0122] The attributes of the item to be displayed can determine the required display area or suitable display location for the item, thereby determining the placement area of ​​the item to be displayed by combining the specific attributes of the pedestrian eye-gathering area.

[0123] It should be noted that in practical applications, the area of ​​the pedestrian gaze cluster is not required to be larger than the display area required for the item to be displayed. For example, if the pedestrian gaze cluster can accommodate the core part of the item to be displayed, and the surrounding area of ​​the pedestrian gaze cluster can accommodate the item to be displayed, then the pedestrian gaze cluster can be used as the placement area for the item to be displayed.

[0124] In this embodiment of the invention, the purpose of utilizing the pedestrian sight gathering area is to ensure that the location of the item to be displayed is more easily seen by pedestrians. During the actual placement of the item to be displayed, it is still necessary to determine the placement area based on the actual situation of the display area. By applying the technical solution of the present invention, the item to be displayed can be placed in a more reasonable position based on the pedestrian sight gathering area.

[0125] Figure 5 A flowchart of yet another embodiment of a method for determining a pedestrian line-of-sight convergence zone provided by the present invention is shown. This method for determining a pedestrian line-of-sight convergence zone is performed by a device for determining a pedestrian line-of-sight convergence zone. Figure 5 As shown, the method for determining the pedestrian line-of-sight convergence zone includes the following steps:

[0126] Step 510: Obtain the pedestrian image of the area to be tested and the area model corresponding to the area to be tested.

[0127] Step 520: Based on the pedestrian images, estimate the line of sight for each pedestrian.

[0128] Specifically, based on the pedestrian images, the gaze of each pedestrian can be estimated using a gaze recognition model. Specifically, the pedestrian images are input into a pre-trained gaze recognition model, which outputs the pedestrian's head pose. Then, based on the pedestrian's head pose and the region model, the pedestrian's gaze is estimated.

[0129] It should be noted that the gaze recognition model is a model built using deep learning algorithms, with the core algorithm being img2pose. This core algorithm can directly estimate the 6 degrees of freedom of the pedestrian's head in a pedestrian image. For the task of estimating pedestrian gaze, it is only necessary to obtain the orientation of the pedestrian's head in the world coordinate system.

[0130] Step 530: Based on the estimated pedestrian line of sight and the area model, determine the effective pedestrian line of sight for each pedestrian in the display area.

[0131] Since the human eye has a certain viewing distance, and the viewing distance is related to the pedestrian's environment, for example, it can be determined based on the actual situation of the surrounding passageway area of ​​the display area, or based on the lighting brightness and average pedestrian density of the surrounding passageway area, or based on the actual height of the pedestrian.

[0132] By using the area model, the visual distance of the human eye is confirmed, and combined with the estimated line of sight of each pedestrian, the effective line of sight of each pedestrian in the display area is determined, ensuring a more accurate determination of the pedestrian's line of sight.

[0133] It should be noted that in actual scenarios, the display area has certain boundaries, which can be divided into two types: physical boundaries and virtual boundaries. For the case of physical boundaries, there are transparent boundaries and non-transparent boundaries. That is, the display area includes transparent boundaries and non-transparent boundaries. Therefore, it is necessary to obtain the boundaries of the display area. If the pedestrian's line of sight points to the non-transparent boundary, it means that the pedestrian's line of sight cannot fall on the display area. In this case, it is necessary to remove the line of sight that corresponds to the estimated pedestrian line of sight.

[0134] Specifically, the pedestrian line of sight includes the direction of sight. First, based on the area model, the transparent boundary within the display area is determined. Then, based on the transparent boundary and the direction of sight, the transparent pedestrian line of sight for each pedestrian is determined. The transparent pedestrian line of sight includes the pedestrian's line of sight for observing the display area. Finally, based on the transparent pedestrian line of sight and a preset line of sight observation threshold, the effective pedestrian line of sight for each pedestrian corresponding to the transparent boundary is determined.

[0135] In practical applications, the line-of-sight observation threshold can be directly set to the human eye's visual distance, or it can be preset based on the eye's visual distance and a region model.

[0136] Step 540: Mark the intersection of the effective pedestrian lines of sight of all pedestrians as the pedestrian line of sight intersection points in the display area.

[0137] The intersection of the effective pedestrian lines of sight for all pedestrians refers to the intersection formed by the mutual intersection of pedestrian lines of sight that can fall into the display area. In practical applications, even if a pedestrian's line of sight points to a certain area, it may not fall into the display area due to obstacles or the influence of the area. Therefore, it is necessary to determine the pedestrian lines of sight that can fall into the display area based on the pedestrian image and the area model.

[0138] Step 550: Determine the pedestrian gaze concentration area within the display area based on the density of the pedestrian gaze intersection points within the display area.

[0139] By applying the technical solution of the present invention, interference from the boundary of the display area can be effectively removed, and the line of sight of pedestrians in the display area can be obtained more accurately, further reducing the misjudgment of the line of sight of pedestrians in the display area.

[0140] It should be noted that in practical applications, the display area may have one or more transparent boundaries. If the display area includes one transparent boundary, the pedestrian gaze concentration area within the display area is determined based on the density of the intersection points of the pedestrian gaze corresponding to the transparent boundary. If the display area includes two or more transparent boundaries, in order to ensure that the determined pedestrian gaze concentration area is more conducive to placing the displayed items, this embodiment of the application can calculate the pedestrian gaze concentration area of ​​each transparent boundary separately, and take the intersection of the pedestrian gaze concentration areas of all transparent boundaries as the final pedestrian gaze concentration area. In this way, it is ensured that pedestrians in the surrounding passage area can observe the final concentration area from any of the transparent boundaries, thereby enabling the displayed items placed in the final pedestrian gaze concentration area to attract more pedestrian gaze.

[0141] In summary, this invention provides a method for determining pedestrian gaze concentration zones. First, pedestrian images of the area to be tested and a corresponding region model are acquired. Then, based on the pedestrian images and the region model, pedestrian gaze intersection points within the display area are determined. Finally, the density of these pedestrian gaze intersection points within the display area is used to determine the pedestrian gaze concentration zone. Applying this invention allows for a more scientific determination of pedestrian gaze concentration zones in surrounding passageways within the display area. This enables the placement of items to be displayed within these zones, ensuring they are seen by more pedestrians and attracting their attention.

[0142] Figure 6 A schematic diagram of an embodiment of a pedestrian line-of-sight determination device provided by the present invention is shown. Figure 6 As shown, the pedestrian gaze concentration area determination device 600 includes: a first acquisition module 610, a gaze intersection point determination module 620, a pedestrian gaze concentration area determination module 630, a second acquisition module 640, and a placement area generation module 650.

[0143] The first acquisition module 610 is used to acquire pedestrian images of the area to be tested and the area model corresponding to the area to be tested.

[0144] The line-of-sight intersection determination module 620 is used to determine the line-of-sight intersection points of pedestrians within the display area based on the pedestrian image and the region model.

[0145] The pedestrian gaze concentration area determination module 630 is used to determine the pedestrian gaze concentration area in the display area based on the density of the pedestrian gaze intersection points in the display area.

[0146] In one implementation, the second acquisition module 640 is used to acquire the attributes of the items to be displayed in the display area, the attributes of the items to be displayed including at least one of the following: the area occupied by the items to be displayed, the accessible range for pedestrians, or the safe distance between the items to be displayed and surrounding items.

[0147] The placement area generation module 650 is used to generate a placement area for the item to be displayed based on the attributes of the item to be displayed and the pedestrian eye focus area.

[0148] In one implementation, the line-of-sight intersection determination module 620 is further configured to:

[0149] Based on the pedestrian images, the line of sight of each pedestrian is estimated.

[0150] Based on the estimated pedestrian line of sight and the area model, the effective pedestrian line of sight for each pedestrian in the display area is determined.

[0151] Mark the intersection of the effective pedestrian lines of sight of all pedestrians as the pedestrian line of sight intersection point within the display area.

[0152] In one implementation, the pedestrian line-of-sight determination module 620 is further configured to:

[0153] Obtain the boundary of the display area, which includes a first boundary that is visible to pedestrians and a second boundary that is not visible to pedestrians.

[0154] If the direction of any pedestrian's gaze is pointing towards the second boundary, then the gaze of any pedestrian corresponding to that gaze direction is removed.

[0155] In one implementation, the pedestrian line-of-sight determination module 620 is further configured to:

[0156] The pedestrian image is input into a pre-trained gaze recognition model, which outputs the pedestrian's head posture.

[0157] Based on the pedestrian's head posture and the region model, the pedestrian's line of sight is predicted.

[0158] In one implementation, the step of determining the pedestrian gaze concentration area within the display area based on the density of pedestrian gaze intersections within the display area specifically includes:

[0159] If the display area includes one transparent boundary, then the pedestrian gaze concentration area within the display area is determined based on the density of the pedestrian gaze intersection points corresponding to the transparent boundary; if the display area includes two or more transparent boundaries, then the pedestrian gaze concentration area of ​​each transparent boundary is calculated based on the density of the pedestrian gaze intersection points corresponding to each transparent boundary, and the intersection of the pedestrian gaze concentration areas of all transparent boundaries is taken as the final pedestrian gaze concentration area.

[0160] This invention provides a device for determining pedestrian gaze concentration zones. First, a first acquisition module 610 acquires pedestrian images of the area to be tested and a corresponding region model. Then, a gaze intersection point determination module 630 determines pedestrian gaze intersection points within the display area based on the pedestrian images and the region model. Finally, a pedestrian gaze concentration zone determination module 670 determines the pedestrian gaze concentration zone within the display area based on the density of the pedestrian gaze intersection points. Applying this invention's technical solution allows for a more scientific determination of pedestrian gaze concentration zones in surrounding passageways within the display area. This allows items to be displayed to be placed within these zones, ensuring they are seen by more pedestrians and attracting their attention.

[0161] Figure 7 The diagram shows a structural schematic of an embodiment of a device for determining pedestrian visual focus areas provided by the present invention. The specific embodiments of the present invention do not limit the specific implementation of the device for determining pedestrian visual focus areas.

[0162] like Figure 7 As shown, the device for determining the pedestrian gaze area may include: a processor 702, a communications interface 704, a memory 706, and a communications bus 708.

[0163] The processor 702, communication interface 704, and memory 706 communicate with each other via communication bus 708. Communication interface 704 is used to communicate with other network elements such as clients or other servers. Processor 702 executes program 710, specifically performing the relevant steps described in the embodiment of the method for determining pedestrian line-of-sight clusters.

[0164] Specifically, program 710 may include program code, which includes computer-executable instructions.

[0165] The processor 702 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The device for determining the pedestrian gaze area may include one or more processors of the same type, such as one or more CPUs; or it may include processors of different types, such as one or more CPUs and one or more ASICs.

[0166] Memory 706 is used to store program 710. Memory 706 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0167] Specifically, program 710 can be called by processor 702 to cause the device determining the pedestrian gaze focus area to perform the following operations:

[0168] Obtain pedestrian images of the area to be tested and the corresponding region model of the area to be tested.

[0169] Based on the pedestrian images and the region model, the pedestrian line-of-sight intersection points within the display area are determined.

[0170] Based on the density of the pedestrian line-of-sight intersections within the display area, the pedestrian line-of-sight concentration area within the display area is determined.

[0171] In one implementation, after the step of determining the pedestrian sight-gathering area within the display area based on the density of the sight intersections, the method further includes:

[0172] Obtain the attributes of the items to be displayed in the display area, wherein the attributes of the items to be displayed include at least one of the following: the area occupied by the items to be displayed, the accessible range for pedestrians, or the safe distance between the items to be displayed and surrounding items.

[0173] Based on the attributes of the item to be displayed and the pedestrian eye focus area, a placement area for the item to be displayed is generated.

[0174] In one implementation, the step of determining the intersection point of pedestrian gazes within the display area based on the pedestrian image and the region model specifically includes:

[0175] Based on the pedestrian images, the line of sight of each pedestrian is estimated.

[0176] Based on the estimated pedestrian line of sight and the area model, the effective pedestrian line of sight for each pedestrian in the display area is determined.

[0177] Mark the intersection of the effective pedestrian lines of sight of all pedestrians as the pedestrian line of sight intersection point within the display area.

[0178] In one implementation, the pedestrian gaze includes the gaze direction; the step of determining the effective pedestrian gaze of each pedestrian within the display area based on the estimated pedestrian gaze and the region model specifically includes:

[0179] Based on the area model, determine the transparent boundaries within the display area;

[0180] Based on the transparent boundary and the direction of sight, determine the transparent pedestrian sightlines of all pedestrians; the transparent pedestrian sightlines include: the sightline information of the pedestrians' transparent observation display area;

[0181] Based on the visible pedestrian line of sight and the preset line of sight observation threshold, the effective pedestrian line of sight for each pedestrian corresponding to the visible boundary is determined.

[0182] In one implementation, the step of estimating the pedestrian's gaze based on the pedestrian image further includes:

[0183] The pedestrian image is input into a pre-trained gaze recognition model, which outputs the pedestrian's head posture.

[0184] Based on the pedestrian's head posture and the region model, the pedestrian's line of sight is predicted.

[0185] In one implementation, the step of determining the pedestrian gaze concentration area within the display area based on the density of pedestrian gaze intersections within the display area specifically includes:

[0186] If the display area includes a transparent boundary, then the pedestrian gaze concentration area within the display area is determined based on the density of the pedestrian gaze intersection points corresponding to the transparent boundary.

[0187] If the display area includes two or more transparent boundaries, then according to the density of the pedestrian line-of-sight intersection points corresponding to each transparent boundary, the pedestrian line-of-sight gathering area of ​​each transparent boundary is calculated, and the intersection of the pedestrian line-of-sight gathering areas of all transparent boundaries is taken as the final pedestrian line-of-sight gathering area.

[0188] This invention provides a device for determining pedestrian gaze concentration areas. The device's memory 706 stores a program 710, which can be implemented by a processor 702. The program first acquires pedestrian images of the area to be tested and a corresponding region model. Then, based on the pedestrian images and the region model, it determines pedestrian gaze intersections within the display area. Finally, it determines the pedestrian gaze concentration area within the display area based on the density of these intersections. Applying this invention allows for a more scientific determination of pedestrian gaze concentration areas in surrounding passageways within the display area. This enables the placement of displayed items within these areas, ensuring they are seen by more pedestrians and attracting their attention.

[0189] This invention provides a computer-readable storage medium storing at least one executable instruction. When the executable instruction is executed on a device / apparatus for determining a pedestrian line-of-sight area, the device / apparatus performs the method for determining a pedestrian line-of-sight area in any of the above-described method embodiments.

[0190] Specifically, the executable instructions can be used to cause the device / organization that determines the pedestrian line of sight area to perform the following operations:

[0191] Obtain pedestrian images of the area to be tested and the corresponding region model of the area to be tested.

[0192] Based on the pedestrian images and the region model, the pedestrian line-of-sight intersection points within the display area are determined.

[0193] Based on the density of the pedestrian line-of-sight intersections within the display area, the pedestrian line-of-sight concentration area within the display area is determined.

[0194] In one implementation, after the step of determining the pedestrian sight-gathering area within the display area based on the density of the sight intersections, the method further includes:

[0195] Obtain the attributes of the items to be displayed in the display area, wherein the attributes of the items to be displayed include at least one of the following: the area occupied by the items to be displayed, the accessible range for pedestrians, or the safe distance between the items to be displayed and surrounding items.

[0196] Based on the attributes of the item to be displayed and the pedestrian eye focus area, a placement area for the item to be displayed is generated.

[0197] In one implementation, the step of determining the intersection point of pedestrian gazes within the display area based on the pedestrian image and the region model specifically includes:

[0198] Based on the pedestrian images, the line of sight of each pedestrian is estimated.

[0199] Based on the estimated pedestrian line of sight and the area model, the effective pedestrian line of sight for each pedestrian in the display area is determined.

[0200] Mark the intersection of the effective pedestrian lines of sight of all pedestrians as the pedestrian line of sight intersection point within the display area.

[0201] In one implementation, the pedestrian gaze includes the gaze direction; the step of determining the effective pedestrian gaze of each pedestrian within the display area based on the estimated pedestrian gaze and the region model specifically includes:

[0202] Based on the area model, determine the transparent boundaries within the display area;

[0203] Based on the transparent boundary and the direction of sight, determine the transparent pedestrian sightlines of all pedestrians; the transparent pedestrian sightlines include: the sightline information of the pedestrians' transparent observation display area;

[0204] Based on the visible pedestrian line of sight and the preset line of sight observation threshold, the effective pedestrian line of sight for each pedestrian corresponding to the visible boundary is determined.

[0205] In one implementation, the step of estimating the pedestrian's gaze based on the pedestrian image further includes:

[0206] The pedestrian image is input into a pre-trained gaze recognition model, which outputs the pedestrian's head posture.

[0207] Based on the pedestrian's head posture and the region model, the pedestrian's line of sight is predicted.

[0208] In one implementation, the step of determining the pedestrian gaze concentration area within the display area based on the density of pedestrian gaze intersections within the display area specifically includes:

[0209] If the display area includes a transparent boundary, then the pedestrian gaze concentration area within the display area is determined based on the density of the pedestrian gaze intersection points corresponding to the transparent boundary.

[0210] If the display area includes two or more transparent boundaries, then according to the density of the pedestrian line-of-sight intersection points corresponding to each transparent boundary, the pedestrian line-of-sight gathering area of ​​each transparent boundary is calculated, and the intersection of the pedestrian line-of-sight gathering areas of all transparent boundaries is taken as the final pedestrian line-of-sight gathering area.

[0211] This invention provides a computer-readable storage medium. When the executable instructions stored in the computer-readable storage medium are executed, firstly, pedestrian images of the area to be tested and a corresponding region model of the area to be tested are acquired; then, based on the pedestrian images and the region model, pedestrian eye-tracking intersections within the display area are determined; finally, the density of the pedestrian eye-tracking intersections within the display area is used to determine the pedestrian eye-tracking concentration area within the display area. Applying the technical solution of this invention allows for a more scientific determination of pedestrian eye-tracking concentration areas in the surrounding passageway areas within the display area. This allows items to be displayed to be placed within these concentration areas, ensuring that they are seen by more pedestrians and attracting their attention.

[0212] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Furthermore, the embodiments of this invention are not directed to any particular programming language.

[0213] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. Similarly, for the sake of brevity and to aid in understanding one or more aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0214] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

[0215] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A method for determining the pedestrian line-of-sight convergence zone, characterized in that, A method for determining a pedestrian visual line gathering area in a to-be-tested region, the to-be-tested region comprising a display area and a peripheral passage area, the method comprising: acquiring a pedestrian image of the to-be-tested region and a region model corresponding to the to-be-tested region; determining a pedestrian visual line intersection point in the display area based on the pedestrian image and the region model; determining a pedestrian visual line gathering area in the display area according to a density of the pedestrian visual line intersection points in the display area; wherein the step of determining the pedestrian visual line gathering area in the display area according to the density of the pedestrian visual line intersection points in the display area specifically comprises: if the display area comprises one transparent boundary, determining the pedestrian visual line gathering area in the display area according to the density of the pedestrian visual line intersection points corresponding to the transparent boundary; if the display area comprises more than two transparent boundaries, respectively calculating a pedestrian visual line gathering area of each transparent boundary according to the density of the pedestrian visual line intersection points corresponding to each transparent boundary, and taking an intersection of the pedestrian visual line gathering areas of all the transparent boundaries as a final pedestrian visual line gathering area.

2. The method of determining a line-of-sight concentration zone for a pedestrian according to claim 1, wherein, after the step of determining the pedestrian visual line gathering area in the display area according to the density of the visual line intersection points, the method further comprises: acquiring attributes of a to-be-displayed article in the display area, the attributes of the to-be-displayed article comprising at least one of a to-be-displayed article footprint, a pedestrian contactable range or a safety distance of the to-be-displayed article from a peripheral article; generating a placement area of the to-be-displayed article based on the attributes of the to-be-displayed article and the pedestrian visual line gathering area.

3. The method of claim 1, wherein, the step of determining the pedestrian visual line intersection point in the display area based on the pedestrian image and the region model specifically comprises: estimating a pedestrian visual line of each pedestrian based on the pedestrian image; determining an effective pedestrian visual line of each pedestrian in the display area based on the estimated pedestrian visual line of each pedestrian and the region model; marking an intersection of the effective pedestrian visual lines of all the pedestrians as the pedestrian visual line intersection point in the display area.

4. The method of determining a line-of-sight concentration zone for a pedestrian according to claim 3, wherein, the pedestrian visual line comprises a visual line direction; the step of determining the effective pedestrian visual line of each pedestrian in the display area based on the estimated pedestrian visual line of each pedestrian and the region model specifically comprises: determining a transparent boundary in the display area according to the region model; determining a transparent pedestrian visual line of each pedestrian according to the transparent boundary and the visual line direction, the transparent pedestrian visual line comprising a visual line by which the pedestrian can transparently observe the display area; determining an effective pedestrian visual line of each pedestrian corresponding to the transparent boundary according to the transparent pedestrian visual line and a preset visual line observation threshold.

5. The method of claim 3, wherein, in the step of estimating the pedestrian visual line of each pedestrian based on the pedestrian image, the method further comprises: inputting the pedestrian image into a pre-trained visual line recognition model to output a head posture of the pedestrian; estimating the pedestrian visual line of the pedestrian based on the head posture of the pedestrian and the region model.

6. A device for determining the pedestrian line-of-sight convergence zone, characterized in that, The pedestrian visual line cluster area determination device is used for determining a pedestrian visual line cluster area of a to-be-tested region, the to-be-tested region comprising a display area and a surrounding passage area, and the pedestrian visual line cluster area determination device comprises: a first acquisition module, configured to acquire a pedestrian image of the to-be-tested region and a region model corresponding to the to-be-tested region; a visual line intersection point determination module, configured to determine a visual line intersection point of a pedestrian in the display area based on the pedestrian image and the region model; a pedestrian visual line cluster area determination module, configured to determine a pedestrian visual line cluster area in the display area according to a density of the visual line intersection points of the pedestrian in the display area; the pedestrian visual line cluster area determination module is further configured to, if the display area comprises one transparent boundary, determine the pedestrian visual line cluster area in the display area according to a density of the visual line intersection points corresponding to the transparent boundary; and if the display area comprises more than two transparent boundaries, calculate a pedestrian visual line cluster area of each transparent boundary according to a density of the visual line intersection points corresponding to each transparent boundary, respectively, and take an intersection of the pedestrian visual line cluster areas of all the transparent boundaries as a final pedestrian visual line cluster area.

7. The apparatus for determining a line-of-sight concentration zone of a pedestrian according to claim 6, wherein The pedestrian visual line cluster area determination device further comprises: a second acquisition module, configured to acquire an attribute of a to-be-displayed article in the display area, the attribute of the to-be-displayed article comprising at least one of a footprint of the to-be-displayed article, a pedestrian contactable range or a safety distance of the to-be-displayed article from surrounding articles; a placement region generation module, configured to generate a placement region of the to-be-displayed article based on the attribute of the to-be-displayed article and the pedestrian visual line cluster area.

8. A device for determining a pedestrian sightline concentration area, characterized by, comprise: a processor, a memory, a communication interface and a communication bus, the processor, the memory and the communication interface being capable of communicating with each other through the communication bus; the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to perform the operations of the pedestrian visual line cluster area determination method according to any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, The storage medium stores at least one executable instruction, and the executable instruction causes the pedestrian visual line cluster area determination device / apparatus to perform the operations of the pedestrian visual line cluster area determination method according to any one of claims 1-5 when the pedestrian visual line cluster area determination device / apparatus runs the executable instruction.

Images