Multi-angle face data acquisition method, device, equipment and storage medium

Abstract

The present disclosure provides a multi-angle facial data acquisition method, apparatus, device, and storage medium, which relates to the field of image processing technology, specifically to the technical fields of computer vision, artificial intelligence, autonomous driving, data acquisition, and facial recognition. A specific implementation scheme is as follows: acquiring a current image acquired by a camera and determining a first facial orientation of a face in the current image; when the first facial orientation is a frontal orientation, acquiring facial feature values ​​of the face in the current image to obtain a frontal feature value; after obtaining the frontal feature value, continuing to acquire images acquired by the camera and determining a second facial orientation of the face in the current image until acquiring facial feature values ​​of a face in an image corresponding to a second facial orientation of at least one side face orientation to obtain at least one side face feature value; based on the frontal feature value and at least one side face feature value, determining and saving the target facial feature value, thereby achieving multi-angle facial data acquisition.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing technology, specifically to technical fields such as computer vision, artificial intelligence, autonomous driving, data acquisition, and face recognition, and more particularly to a multi-angle face data acquisition method, apparatus, device, and storage medium. Background Art

[0002] With the development of computer image processing technology, facial recognition technology has been widely used in various fields. In facial recognition scenarios, the captured face is compared and analyzed with pre-registered faces to achieve recognition. Face registration involves extracting facial features from facial images captured by a camera and linking the extracted features with the information of the person whose face was captured. Summary of the Invention

[0003] The present disclosure provides a multi-angle facial data acquisition method, apparatus, device and storage medium.

[0004] According to one aspect of the present disclosure, a multi-angle face data collection method is provided, comprising:

[0005] Obtaining a current image captured by a camera, and determining a first face orientation corresponding to a posture angle of a face in the current image;

[0006] When the first face orientation is a frontal face orientation, acquiring facial feature values ​​of the face in the current image to obtain frontal face feature values;

[0007] When the frontal face feature value is obtained, continue to obtain images captured by the camera, and determine a second facial orientation corresponding to the posture angle of the face in the currently obtained image, until each of the determined second facial orientations includes at least one side face orientation, and obtain facial feature values ​​of the faces in the images corresponding to the second facial orientations of the at least one side face orientation, to obtain at least one side face feature value;

[0008] Based on the frontal face feature value and the at least one side face feature value, the target face feature value is determined and saved.

[0009] According to another aspect of the present disclosure, a multi-angle face data acquisition device is provided, comprising:

[0010] a first image acquisition module, configured to acquire a current image captured by a camera and determine a first face orientation corresponding to a posture angle of a face in the current image;

[0011] a frontal face feature value acquisition module, configured to acquire facial feature values ​​of the face in the current image when the first face orientation is a frontal face orientation, to obtain frontal face feature values;

[0012] a side face feature value acquisition module, configured to, upon obtaining the frontal face feature value, continue to acquire images captured by the camera, and determine a second facial orientation corresponding to a posture angle of the face in the currently acquired image, until each of the determined second facial orientations includes a side face orientation in at least one direction, and acquire a facial feature value of the face in the image corresponding to the second facial orientation in the at least one direction, thereby obtaining at least one side face feature value;

[0013] A target feature value determination module is used to determine the target facial feature value based on the frontal facial feature value and the at least one side facial feature value, and save the target facial feature value.

[0014] According to another aspect of the present disclosure, there is provided an electronic device, comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any one of the methods described in the present disclosure.

[0018] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to execute any one of the methods of the present disclosure.

[0019] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein the computer program implements any one of the methods of the present disclosure when executed by a processor.

[0020] The disclosed embodiments achieve multi-angle facial data collection.

[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0023] Figure 1is a schematic diagram of the multi-angle face data collection method according to the present disclosure;

[0024] Figure 2 This is a schematic diagram of the face orientation area division according to the present disclosure;

[0025] Figure 3 is a schematic diagram of a process for obtaining side face feature values ​​according to the present disclosure;

[0026] Figure 4 is a schematic diagram of the multi-angle face registration method according to the present disclosure;

[0027] Figure 5 This is a schematic diagram of a multi-angle face data acquisition device according to the present disclosure;

[0028] Figure 6 It is a block diagram of an electronic device used to implement the multi-angle face data collection method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0029] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0030] With the development of computer image processing technology, face recognition technology has been widely used in various fields. During face recognition, a camera captures a user's facial image and detects facial information on the image to obtain the user's facial feature values. These are then compared and analyzed with the facial feature values ​​pre-registered in the face recognition system to achieve face recognition. To ensure the accuracy of face recognition, related technical solutions, during the face registration process, guide the user to face the camera as closely as possible through the human-computer interaction interface. While ensuring a clear and complete display of the face, the face image is captured and facial feature values ​​are extracted from the face image. The extracted facial feature values ​​are associated with the user and stored in the face recognition system to complete face registration. During face recognition, the user is also required to face the camera as closely as possible to collect the frontal facial feature values. The collected frontal facial feature values ​​are then compared and analyzed with the registered facial feature values ​​to achieve face recognition.

[0031] Because the facial feature values ​​corresponding to the face facing the camera are collected during face registration, when performing face recognition, the user needs to keep the face angle close to the one at which the facial feature values ​​were collected during face registration. In this way, during the face recognition process, the frontal facial feature values ​​with high similarity to the facial feature values ​​collected during face registration can be obtained, so as to accurately obtain the recognition results.

[0032] However, during face recognition using the above-mentioned technical solution, if the user's face registration image quality is poor or the user's face is not facing the camera during face registration, the facial feature values ​​stored in the face recognition system are inaccurate, that is, they cannot accurately describe the feature information of the user's face. As a result, the accuracy of subsequent face recognition will be significantly reduced. For example, when user A is performing face registration, because the user's face is not facing the camera, the collected facial feature values ​​cannot accurately describe the feature information of the user's face. As a result, when the face recognition system performs face recognition on user A, it may repeatedly prompt user A to change the face angle and face the camera before trying again because it cannot match the registered facial feature values, or it may mistakenly identify user A as user B. The above-mentioned technical solution requires that the face angle during face recognition be as similar as possible to the face angle during face registration, and the quality of the face recognition image captured by the camera must be highly restored to the image quality during face registration. In addition, face recognition is easily affected by factors such as the user's degree of coordination and the quality of the image captured by the camera, resulting in unstable recognition results.

[0033] To solve the above problems, the present disclosure provides a multi-angle face data collection method, comprising:

[0034] Obtaining a current image captured by a camera, and determining a first face orientation corresponding to a posture angle of a face in the current image;

[0035] When the first face orientation is a frontal face orientation, acquiring facial feature values ​​of the face in the current image to obtain frontal face feature values;

[0036] When the frontal face feature value is obtained, continue to obtain images captured by the camera, and determine a second facial orientation corresponding to the posture angle of the face in the currently obtained image, until each of the determined second facial orientations includes at least one side face orientation, and obtain facial feature values ​​of the faces in the images corresponding to the second facial orientations of the at least one side face orientation, to obtain at least one side face feature value;

[0037] Based on the frontal face feature value and the at least one side face feature value, the target face feature value is determined and saved.

[0038] In the embodiment of the present disclosure, when the first face orientation corresponding to the current image collected is the frontal face orientation, the facial feature values ​​of the face in the current image are obtained to give priority to obtaining the frontal face feature values. When the frontal face feature values ​​are obtained, further, the facial feature values ​​corresponding to the second face orientation in the collected image that is the side face orientation in at least one direction are obtained to obtain at least one side face feature value, thereby expanding the angle range of facial data collection. Then, based on the frontal face feature value and at least one side face feature value, the target facial feature value that fuses the frontal face feature value and at least one side face feature value is obtained, thereby realizing multi-angle facial data collection and improving the integrity of facial data collection.

[0039] The multi-angle facial data collection method provided by the embodiments of the present disclosure can be applied to scenarios such as facial registration, facial data collection, and facial recognition. In one example, it can be applied to scenarios such as facial registration and facial recognition with a single camera in a fixed position. It can also be applied to products such as the Image Delivery Baseline Software Development Kit for Adaptive Software Development Engineering Delivery Centers. The multi-angle facial data collection method provided by the embodiments of the present disclosure can be applied to electronic devices such as servers, smart terminals, and the like.

[0040] The multi-angle face data collection method provided by the embodiment of the present disclosure is described in detail below.

[0041] See also Figure 1 The embodiment of the present disclosure provides a multi-angle face data collection method, comprising the following steps:

[0042] S110: Acquire a current image captured by a camera, and determine a first face orientation corresponding to a posture angle of a face in the current image.

[0043] In an example, a single camera at a fixed position can be used to capture content within the coverage area of ​​the camera to obtain the current image. The single camera at a fixed position can be a car camera, or any other camera in any application scenario that requires facial data collection.

[0044] After obtaining the current image, the face detection model can be used to detect the faces, face posture angles, etc. contained in the current image. Different face posture angles correspond to different face orientations. After determining the face posture angle, the face orientation corresponding to the face posture angle can be determined based on the relationship between the face posture angle and the face orientation.

[0045] In a possible implementation, the determining of the first face orientation corresponding to the posture angle of the face in the current image may include the following steps:

[0046] Step 1: Perform face detection on the current image to obtain face information in the current image, wherein the face information includes the target posture angle of the face.

[0047] Step 2: Based on the correspondence between the preset posture angle and the face orientation, determine the first face orientation corresponding to the target posture angle, where the face orientation includes the frontal face orientation and the side face orientation in multiple directions.

[0048] In one example, a face detection model can be used to perform face detection on a current image to obtain a target pose angle of the face in the current image. The face detection model can be trained based on sample images and the pose angles of the faces in the sample images.

[0049] In the embodiment of the present disclosure, since different cameras correspond to different parameters of the image acquisition device, different correspondences between posture angles and face orientations can be pre-set for different cameras. Then, when the target posture angle of the face in the current image is detected, the correspondence can be queried to determine the first face orientation corresponding to the target posture angle.

[0050] In the embodiment of the present disclosure, face detection is performed on the current image to obtain the target posture angle of the face in the current image, and then the first face orientation corresponding to the target posture angle is accurately determined based on the correspondence between the preset posture angle and the face orientation, so as to facilitate the accurate collection of facial feature values ​​under different face orientations.

[0051] In a possible implementation, the posture angle may include a pitch angle and a yaw angle. Accordingly, the correspondence between the preset posture angle and the face orientation includes: the mapping relationship between each preset pitch angle threshold interval and yaw angle threshold interval and each face orientation.

[0052] In the disclosed embodiment, the image captured by the camera is divided into a plurality of face orientation areas according to the pitch (pitch angle) / yaw (yaw angle) coordinate system, and different face orientations correspond to different pitch angle threshold intervals and yaw angle threshold intervals.

[0053] In one example, because the trajectory of a face rotating in different directions is approximately an elliptical trajectory, which is a biological characteristic, and different cameras have different imaging principles and parameters, the area corresponding to the frontal face orientation (such as the area contained in the elliptical trajectory) can be prioritized in the pitch / yaw coordinate system based on the camera imaging conditions and the biological characteristics of facial deflection. Then, based on the camera edge imaging conditions and empirical values ​​(such empirical values ​​can be obtained by manually dividing the test images into regions using a certain number of test images collected with the camera), the pitch angle threshold range and yaw angle threshold range corresponding to each face orientation are determined, and multiple face orientation regions are determined in the pitch / yaw coordinate system. The face orientation regions include a frontal face orientation region and multiple side face orientation regions, with each face orientation region corresponding to each face orientation.

[0054] For example, Figure 2 As shown, the trajectory of a face deflecting in all directions is approximately an ellipse. The ellipse's trajectory parameters can be determined based on the camera type, camera position, and the distance between the camera and the face. The area within the ellipse represents the frontal face orientation. After determining the ellipse's trajectory parameters, the coordinates of points A and B on the ellipse can be determined based on the camera's edge imaging conditions and empirical values. Based on the coordinates of points A and B and the preset face deflection angle extremes (Max yaw, Min yaw, Max pitch, and Min pitch), the camera imaging area can be divided into nine face orientation zones. Figure 2The figure shows the division of each face orientation area when the coordinate values ​​of points A and B are determined in the pitch / yaw coordinate system. The oval area in the middle represents the front face orientation area, and the up, down, left, right, upper left, upper right, lower left, and lower right correspond to the eight side face orientation areas in different directions. Specifically, when the coordinate values ​​of points A and B are determined, the symmetrical points of points A and B along the pitch / yaw direction are determined on the ellipse, and the symmetrical points (y1, -p1), (-y1, p1), and (-y1, -p1) of the coordinate point A (y1, p1) are obtained, and the symmetrical points (y2, -p2), (-y2, p2), and (-y2, -p2) of the coordinate point B (y2, p2) are obtained. Then, the interior of the ellipse is determined as the frontal face facing area, and the upper side face facing area is determined according to the coordinates (y2, p2) of point B, the symmetrical point (-y2, p2), and the preset Max pitch. The lower side face facing area is determined according to the symmetrical points (y2, -p2), (-y2, -p2), and the preset Min pitch. The left face orientation area is determined by the symmetrical points (-y1, p1) and (-y1, -p1) and the preset Min yaw. The right face orientation area is determined by the coordinates of point A (y1, p1) and point B (y2, p2) and the preset Max yaw and Max pitch. The upper left face orientation area is determined by the symmetrical points (-y1, p1) and (-y2, p2) and the preset Max pitch and Min yaw. The lower left face orientation area is determined by the symmetrical points (y1, -p1) and (y2, -p2) and the preset Max yaw and Min pitch. The lower right face orientation area is determined by the symmetrical points (-y1, -p1) and (-y2, -p2) and the preset Min yaw and Min pitch. In other words, the pitch angle threshold range and yaw angle threshold range corresponding to each face orientation are determined based on the coordinates of points A and B and the Max yaw, Min yaw, Max pitch, and Min pitch values.

[0055] In an example, the coordinates of points A and B can be adjusted to change the determination range of different face orientation areas.

[0056] Accordingly, the above step 2 determines the first face orientation corresponding to the target posture angle based on the correspondence between the preset posture angle and the face orientation, including:

[0057] The target pitch angle and the target yaw angle in the target attitude angle are matched with the preset pitch angle threshold interval and the yaw angle threshold interval respectively to determine the first face orientation corresponding to the target attitude angle.

[0058] For example, see Figure 2 , the target pitch angle and target yaw angle are mapped to a coordinate point in the pitch / yaw coordinate system, and this coordinate point is matched with each preset pitch angle threshold interval and yaw angle threshold interval to obtain the first facial orientation corresponding to the target attitude angle. Assuming that the target pitch angle and target yaw angle correspond to the coordinates (a, b) in the pitch / yaw coordinate system, and the result of matching with the pitch angle threshold interval and yaw angle threshold interval corresponding to each facial orientation region is: a>y1, and -p1<b<p1, then the first facial orientation corresponding to the target attitude angle can be determined to be the left facial orientation.

[0059] By matching the target pitch angle and target yaw angle in the target attitude angle with the pitch angle threshold interval and yaw angle threshold interval corresponding to the preset face orientation area, the face orientation area corresponding to the target attitude angle can be determined more accurately.

[0060] See also Figure 1 S120, when the first face orientation is a frontal face orientation, obtain facial feature values ​​of the face in the current image to obtain frontal face feature values.

[0061] In an example, face detection is performed on the current image captured by the camera to determine the first face orientation corresponding to the posture angle of the face in the current image. When the first face orientation is the frontal orientation, the face feature value of the face in the current image is preferentially obtained to obtain the frontal face feature value.

[0062] For example, after determining the first facial orientation corresponding to the posture angle of the face in the current image, you can first determine whether the first facial orientation is a frontal orientation. When the first facial orientation is a frontal orientation, obtain the facial feature value of the face in the current image, obtain the frontal feature value, and cache the frontal feature value.

[0063] S130, when the frontal face feature value is obtained, continue to obtain images captured by the camera, and determine the second face orientation corresponding to the posture angle of the face in the currently obtained image, until each determined second face orientation includes at least one direction of side face orientation, obtain the face feature value of the face in the image corresponding to the second face orientation of at least one direction of side face orientation, and obtain at least one side face feature value.

[0064] When the frontal face feature value is obtained, continue to obtain images captured by the camera, and perform face detection on the currently obtained image to determine the second face orientation corresponding to the posture angle of the face in the currently obtained image, until a second face orientation that is the same as the side face orientation in at least one direction is obtained, obtain the face feature value of the image corresponding to the second face orientation, and obtain at least one side face feature value.

[0065] S140: Determine a target facial feature value based on the frontal facial feature value and at least one profile facial feature value, and save the target facial feature value.

[0066] In one example, the frontal face feature value and at least one side face feature value can be fused, for example, the sum, concatenation, average, etc. of the frontal face feature value and at least one side face feature value are calculated to obtain the target face feature value, and the target face feature value is saved.

[0067] In a possible implementation, a weighted average of the frontal face feature value and at least one profile face feature value may be determined as the target facial feature value, and the target facial feature value may be saved.

[0068] For example, after obtaining the front face feature value and at least one side face feature value, for example, 1 front face feature value and 4 side face feature values, the front face feature value and the 4 side face feature values ​​can be weighted averaged to obtain the target face feature value, where the weight of the front face feature value is 0.4 and the weight of each side face feature value is 0.15. Specifically, the weight of the front face feature value and the weight of the side face feature value can be set according to actual needs, and the embodiments of the present disclosure are not limited to this.

[0069] By fusing the frontal face eigenvalues ​​and at least one profile face eigenvalue, facial eigenvalues ​​from multiple different face orientations can be combined and stored together to reduce storage space. The fused target face eigenvalues ​​contain facial feature information describing a wider range of face orientations, while also minimizing the impact of low-quality individual images on face recognition during face registration.

[0070] In the embodiment of the present disclosure, when the first face orientation corresponding to the current image collected is the frontal face orientation, the facial feature values ​​of the face in the current image are obtained to give priority to obtaining the frontal face feature values. When the frontal face feature values ​​are obtained, further, the facial feature values ​​corresponding to the second face orientation in the collected image that is the side face orientation in at least one direction are obtained to obtain at least one side face feature value, thereby expanding the angle range of facial data collection. Then, based on the frontal face feature value and at least one side face feature value, the target facial feature value that fuses the frontal face feature value and at least one side face feature value is obtained, thereby realizing multi-angle facial data collection and improving the integrity of facial data collection. Moreover, because multi-angle facial data collection is carried out during the process of facial feature value collection, the obtained target facial feature values ​​contain facial feature information under a wider range of facial posture angles. Therefore, during face recognition, there is no need to strictly require the user to face the camera as much as possible. Face recognition can also be performed by providing a side face image or facing the camera with the side face. This reduces the situation where recognition cannot be caused by the difference between the face angle during face recognition and the face angle during face registration, and reduces the user's operation of adjusting the face angle and retrying, further reducing the face recognition time and improving the accuracy of face recognition and the stability of the recognition effect.

[0071] In a possible implementation, the above method may further include:

[0072] When the first face orientation is not the frontal face orientation, the process returns to the step of obtaining a current image captured by the camera and determining the first face orientation corresponding to the posture angle of the face in the current image.

[0073] In one example, after determining the first face orientation corresponding to the posture angle of the face in the current image, first determine whether the first face orientation is a frontal orientation. If the first face orientation is a frontal orientation, obtain the facial feature value of the face in the current image to obtain the frontal feature value; if the first face orientation is not a frontal orientation, prompt the user to face the camera in the human-computer interaction interface, continue to obtain the current image captured by the camera, and determine the first face orientation corresponding to the posture angle of the face in the current image until the first face orientation is a frontal orientation, so as to preferentially obtain the frontal feature value.

[0074] Prioritize the extraction of frontal face feature values ​​so that side face feature values ​​belonging to the same user as the frontal face feature values ​​can be better collected based on the frontal face feature values.

[0075] In one possible implementation, Figure 3 As shown, the process of obtaining the above-mentioned profile feature value may include:

[0076] S310: When the frontal face feature value is obtained, an image captured by the camera is obtained, and a second face orientation corresponding to the posture angle of the face in the currently obtained image is determined.

[0077] In one example, after acquiring frontal facial feature values, the user can be guided in the human-computer interaction interface to change the facial orientation and continue acquiring images captured by the camera to obtain at least one profile facial feature value. Specifically, the implementation process of determining the second facial orientation corresponding to the posture angle of the face in the currently acquired image in this step can refer to the implementation process of determining the first facial orientation corresponding to the posture angle of the face in the currently acquired image in step S110 above, and will not be further described in this embodiment of the present disclosure.

[0078] S320: Determine whether the second face orientation corresponds to the side face orientation in at least one direction.

[0079] Determine whether the second face orientation hits any of the above 8 side face orientations. If the second face orientation does not correspond to the side face orientation in any direction, it means that the face orientation in the currently collected image is a frontal orientation or the image clarity is low and the face orientation is not recognized, or the image does not contain a face. At this time, execute step S330, otherwise execute step S340.

[0080] S330, when the second face orientation does not correspond to the side face orientation in any direction, return to the execution step: obtain the image captured by the camera, and determine the second face orientation corresponding to the posture angle of the face in the currently acquired image.

[0081] S340: When the second facial orientation corresponds to the side facial orientation in at least one direction, determine whether a side facial feature value corresponding to the side facial orientation of the second facial orientation has been obtained.

[0082] S350, when the side face feature value corresponding to the second face orientation has not been obtained, obtain the facial feature value of the face in the currently obtained image to obtain the side face feature value corresponding to the second face orientation.

[0083] S360, when the side face feature value under the side face orientation corresponding to the second face orientation has been obtained, return to the execution step: obtain the image captured by the camera, and determine the second face orientation corresponding to the posture angle of the face in the currently obtained image, until the side face feature value corresponding to the side face orientation in the preset number of directions is obtained.

[0084] In one example, when the second facial orientation corresponds to at least one side facial orientation, a further determination is made as to whether a side facial feature value corresponding to the second facial orientation has already been obtained. If a side facial feature value corresponding to the second facial orientation has already been obtained, the user can be guided in the human-computer interaction interface to change the facial orientation and reacquire images captured by the camera until side facial feature values ​​corresponding to a preset number of side facial orientations are obtained. The preset number can be set according to actual needs. For example, the preset number of directions can be any number of the directions corresponding to the eight side facial orientation areas described above.

[0085] In this embodiment, by determining the second facial orientation corresponding to the posture angle of the face in the currently acquired image, and determining whether the second facial orientation corresponds to the side facial orientation in at least one direction, and further judging whether the side facial feature value under the side facial orientation corresponding to the second facial orientation has been obtained, repeated extraction of side facial feature values ​​under the same side facial orientation can be avoided, thereby improving the integrity of facial data collection.

[0086] In a possible implementation, the step of obtaining facial feature values ​​of a face in the currently acquired image to obtain the side facial feature values ​​corresponding to the side facial orientation of the second face includes:

[0087] Obtaining facial feature values ​​of the face in the currently acquired image to obtain candidate profile feature values;

[0088] Calculate the similarity between the candidate profile feature value and the frontal face feature value;

[0089] When the similarity value is greater than a preset threshold, the candidate profile feature value is used as the profile feature value under the profile orientation corresponding to the second face orientation and cached;

[0090] If the similarity value is not greater than the preset threshold, the process returns to the step of acquiring an image captured by the camera and determining a second face orientation corresponding to the face posture angle in the currently acquired image.

[0091] In the process of obtaining side face feature values ​​under different side face orientations, the obtained side face feature values ​​under side face orientations are used as candidate side face feature values, and the similarity value between the candidate side face feature value and the front face feature value is further calculated. When the similarity value is greater than a preset threshold, it means that the candidate side face feature value and the front face feature value correspond to the facial feature value of the same user. At this time, the candidate side face feature value is used as the side face feature value under the side face orientation corresponding to the second face orientation and cached; when the similarity value is not greater than the preset threshold, it means that the candidate side face feature value and the front face feature value do not correspond to the facial feature value of the same user. At this time, the user can be guided to change the face orientation in the human-computer interaction interface, and the execution of obtaining the image captured by the camera is returned, and the second face orientation corresponding to the posture angle of the face in the currently acquired image is determined to re-collect the side face feature value under the side face orientation.

[0092] Among them, the similarity value between the candidate side face feature value and the front face feature value can be calculated by calculating the structural similarity value or the cosine similarity value. The preset threshold can be set according to needs, such as 60%, 80% or 90%, etc.

[0093] In the embodiment of the present disclosure, by calculating the similarity value between the side face feature value and the front face feature value to ensure that the front face feature value and the side face feature value are the facial feature values ​​of the same user, the accuracy of facial data collection can be improved.

[0094] In a possible implementation, the facial information may further include facial feature values.

[0095] The facial information also includes: facial feature values, indicating that in the process of face detection on the image, the facial feature values ​​contained in the image have been detected (that is, the facial feature values ​​contained in the image have been extracted). Accordingly, when the first face orientation is the frontal orientation, the facial feature values ​​of the face in the current image are obtained to obtain the frontal feature values. This may include: when the first face orientation is the frontal orientation, the facial feature values ​​of the face in the current image are directly used as the frontal feature values, without the need to detect or extract the facial feature values ​​again.

[0096] The above-mentioned obtaining of the facial feature value of the face in the image with the second face facing in at least one direction of side face orientation, and obtaining at least one side face feature value, may include: determining the facial feature value of the face in the image with the second face facing in at least one direction of side face orientation as the corresponding side face feature value.

[0097] In the embodiment of the present disclosure, during the process of performing face detection on an image, the facial feature values ​​contained in the image are detected. Then, when the face orientation is determined, the facial feature values ​​of the face in the corresponding image are directly determined as the frontal or side face feature values ​​under the face orientation, without the need to detect or extract the facial feature values ​​again.

[0098] In a possible implementation, the facial information further includes: whether there is a face.

[0099] During the process of face detection on an image, it is also possible to detect whether the image contains a face. Furthermore, if the image contains a face, the face orientation corresponding to the posture angle of the face in the current image is determined, and then the face feature values ​​under different face orientations are obtained; if the image does not contain a face, the image captured by the camera is obtained again.

[0100] For example, Figure 4 As shown, Figure 4 FIG. 1 is another schematic diagram of a multi-angle face registration method according to the present disclosure, comprising:

[0101] Step 1: Get the image captured by the camera;

[0102] Step 2: Perform face detection on the image to obtain face information in the image. The face information includes whether there is a face and the target posture angle of the face;

[0103] Step 3: Determine whether there is a face. If no face is found, return to step 1.

[0104] Step 4: When a face is determined to be present, determine the target pose angle of the face in the currently acquired image;

[0105] Step 5: Based on the correspondence between the preset posture angle and the face orientation, determine the target face orientation corresponding to the target posture angle;

[0106] Step 6: Determine whether the frontal face feature value has been obtained;

[0107] Step 7: If it is determined that no frontal face feature value is obtained, further determine whether the target face orientation is a frontal face orientation, and if it is determined that the target face orientation is not a frontal face orientation, return to step 1;

[0108] Step 8: When it is determined that the target face is facing forward, obtain the facial feature value of the face in the currently acquired image, obtain the frontal face feature value, and save the frontal face feature value;

[0109] Step 9: If it is determined that the frontal face feature value has been obtained, further determine whether the target face orientation corresponds to the profile orientation in at least one direction; if it is determined that the target face orientation does not correspond to the profile orientation in any direction, return to step 1;

[0110] Step 10: If it is determined that the target face orientation corresponds to the profile orientation in at least one direction, further determining whether a profile feature value has been obtained for the profile orientation corresponding to the target face orientation; and if it is determined that a profile feature value has been obtained for the profile orientation corresponding to the target face orientation, returning to step 1;

[0111] Step 11: if it is determined that no profile feature values ​​corresponding to the target face orientation have been obtained, facial feature values ​​of the face in the currently obtained image are obtained to obtain candidate profile feature values;

[0112] Step 12: Determine whether the similarity between the candidate profile feature value and the frontal feature value is greater than a preset threshold. If it is determined that the similarity between the candidate profile feature value and the frontal feature value is not greater than the preset threshold, return to step 1.

[0113] Step 13: When it is determined that the similarity between the candidate profile feature value and the frontal face feature value is greater than a preset threshold, the candidate profile feature value is used as the profile feature value for the profile face orientation corresponding to the target face orientation and cached;

[0114] Step 14: Determine whether the preset number of side face feature values ​​in the side face direction have been obtained. If not, return to step 1.

[0115] Step 15: After obtaining the preset number of side face feature values ​​facing downward, determine the target face feature value as a weighted average of the frontal face feature value and the preset number of side face feature values ​​facing downward, and save the target face feature value;

[0116] Step 16: Establish an association between the target facial feature value and the target user to complete face registration.

[0117] In this embodiment, the side face feature values ​​in a preset number of side face directions are all compared with the front face feature values ​​for similarity. When the similarity is greater than a preset threshold, the side face feature values ​​and the front face feature values ​​are determined to be the facial feature values ​​of the same user. The side face feature value extraction of the eight side face regions can be completed according to the above steps. After obtaining the eight side face feature values ​​and one front face feature value of the same user, a feature value fusion algorithm can be used for weighted averaging to merge the nine facial feature values ​​into the target facial feature value. The nine facial feature values ​​can be completely averaged, or different weights can be set for the nine facial feature values ​​for calculation.

[0118] After obtaining the target facial feature values, they can be encrypted and stored in the face recognition system to complete multi-angle facial data collection. Without occupying additional storage space, the target facial feature values ​​can be used to further expand the angle range for facial data collection by increasing the face orientation for facial data collection. This allows facial data to be collected over a wider range of angles, describing facial feature information across a wider range of faces and improving the integrity of facial data collection.

[0119] Based on the same inventive concept, a multi-angle face data acquisition device is provided corresponding to the multi-angle face data acquisition method, such as Figure 5 As shown, the device includes:

[0120] A first image acquisition module 510 is configured to acquire a current image captured by a camera and determine a first face orientation corresponding to a face posture angle in the current image;

[0121] A frontal face feature value acquisition module 520 is configured to acquire facial feature values ​​of the face in the current image when the first face orientation is a frontal face orientation, thereby obtaining frontal face feature values;

[0122] The side face feature value acquisition module 530 is configured to, upon obtaining the frontal face feature value, continue to acquire images captured by the camera, and determine a second facial orientation corresponding to a posture angle of the face in the currently acquired image, until each of the determined second facial orientations includes at least one side face orientation, and then acquire a facial feature value of the face in the image corresponding to the at least one side face orientation in the second facial orientation, thereby obtaining at least one side face feature value.

[0123] The target feature value determination module 540 is used to determine the target facial feature value based on the frontal facial feature value and the at least one profile facial feature value, and save the target facial feature value.

[0124] In the embodiment of the present disclosure, when the first face orientation corresponding to the current image collected is the frontal face orientation, the facial feature values ​​of the face in the current image are obtained to give priority to obtaining the frontal face feature values. When the frontal face feature values ​​are obtained, further, the facial feature values ​​corresponding to the second face orientation in the collected image that is the side face orientation in at least one direction are obtained to obtain at least one side face feature value, thereby expanding the angle range of facial data collection. Then, based on the frontal face feature value and at least one side face feature value, the target facial feature value that fuses the frontal face feature value and at least one side face feature value is obtained, thereby realizing multi-angle facial data collection and improving the integrity of facial data collection. Moreover, because multi-angle facial data collection is carried out during the process of facial feature value collection, the obtained target facial feature values ​​contain facial feature information under a wider range of facial posture angles. Therefore, during face recognition, there is no need to strictly require the user to face the camera as much as possible. Face recognition can also be performed by providing a side face image or facing the camera with the side face. This reduces the situation where recognition cannot be caused by the difference between the face angle during face recognition and the face angle during face registration, and reduces the user's operation of adjusting the face angle and retrying, further reducing the face recognition time and improving the accuracy of face recognition and the stability of the recognition effect.

[0125] In a possible implementation, the determining the first face orientation corresponding to the posture angle of the face in the current image includes:

[0126] Performing face detection on the current image to obtain face information in the current image, wherein the face information includes a target posture angle of the face;

[0127] Based on the correspondence between the preset posture angle and the face orientation, a first face orientation corresponding to the target posture angle is determined; the face orientation includes: a frontal face orientation and a side face orientation in multiple directions.

[0128] In one possible implementation, the above-mentioned attitude angles include: a pitch angle and a yaw angle; the correspondence between the preset attitude angles and the face orientation includes: a mapping relationship between each preset pitch angle threshold interval and yaw angle threshold interval and each face orientation;

[0129] The determining, based on a correspondence between a preset posture angle and a face orientation, a first face orientation corresponding to the target posture angle includes:

[0130] The target pitch angle and the target yaw angle in the target attitude angle are matched with the preset pitch angle threshold interval and the yaw angle threshold interval respectively to determine the first face orientation corresponding to the target attitude angle.

[0131] In a possible implementation, the facial information further includes: facial feature values.

[0132] In a possible implementation, the above device further includes:

[0133] The second image acquisition module is used to trigger the first image acquisition module 510 to execute the following steps when the first face orientation is not the frontal orientation: obtaining the current image captured by the camera and determining the first face orientation corresponding to the posture angle of the face in the current image.

[0134] In a possible implementation, the side face feature value acquisition module 530 includes:

[0135] a first image acquisition unit, configured to acquire an image captured by a camera when the frontal face feature value is obtained, and determine a second face orientation corresponding to a posture angle of the face in the currently acquired image;

[0136] a first determining unit, configured to determine whether the second facial orientation corresponds to a profile orientation in at least one direction;

[0137] a second image acquisition unit configured to, when the first determination unit determines that the second facial orientation does not correspond to the side facial orientation in any direction, trigger the first image acquisition unit to execute: acquiring an image captured by the camera, and determining the second facial orientation corresponding to the posture angle of the face in the currently acquired image;

[0138] a determining unit configured to determine, when the first determining unit determines that the second facial orientation corresponds to a profile orientation in at least one direction, whether a profile feature value corresponding to the profile orientation of the second facial orientation has been obtained;

[0139] a side face feature value acquiring unit, configured to acquire facial feature values ​​of a face in a currently acquired image, and obtain the side face feature value in the side face orientation corresponding to the second face orientation, if the determining unit determines that no side face feature value in the side face orientation corresponding to the second face orientation has been acquired;

[0140] The third image acquisition unit is used to trigger the first image acquisition unit to execute the following steps when the judgment unit determines that the side face feature value under the side face orientation corresponding to the second face orientation has not been obtained: obtaining the image captured by the camera, and determining the second face orientation corresponding to the posture angle of the face in the currently obtained image, until the side face feature values ​​corresponding to the side face orientations in a preset number of directions are obtained.

[0141] In a possible implementation, the step of obtaining facial feature values ​​of a face in the currently acquired image to obtain the side facial feature values ​​corresponding to the side facial orientation of the second face includes:

[0142] Obtaining facial feature values ​​of the face in the currently acquired image to obtain candidate profile feature values;

[0143] Calculating a similarity value between the candidate profile feature value and the frontal face feature value;

[0144] If the similarity value is greater than a preset threshold, the candidate profile feature value is used as a profile feature value corresponding to the profile orientation of the second person and cached;

[0145] If the similarity value is not greater than the preset threshold, the process returns to the step of acquiring an image captured by the camera and determining a second face orientation corresponding to the face posture angle in the currently acquired image.

[0146] In a possible implementation, the target feature value determination module is specifically configured to:

[0147] A weighted average of the frontal face feature value and the at least one side face feature value is determined as the target face feature value, and the target face feature value is saved.

[0148] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0149] The present disclosure provides an electronic device, comprising:

[0150] at least one processor; and

[0151] a memory communicatively connected to at least one processor; wherein,

[0152] The memory stores instructions that can be executed by at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform any method in the present disclosure.

[0153] The present disclosure provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute any one of the methods in the present disclosure.

[0154] The present disclosure provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements any one of the methods in the present disclosure.

[0155] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0156] It should be noted that the head model in this embodiment is not a head model for a specific user and cannot reflect the personal information of a specific user.

[0157] It should be noted that the two-dimensional face images in this embodiment come from a public dataset.

[0158] Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0159] like Figure 6 As shown, the device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. Various programs and data required for the operation of the device 600 can also be stored in the RAM 603. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0160] Various components in device 600 are connected to I / O interface 605, including an input unit 606, such as a keyboard, mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, optical disk, etc.; and a communication unit 609, such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows device 600 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0161] The computing unit 601 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above, such as the multi-angle facial data collection method. For example, in some embodiments, the multi-angle facial data collection method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the multi-angle facial data collection method described above can be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to execute the multi-angle face data acquisition method in any other appropriate manner (for example, by means of firmware).

[0162] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0163] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0164] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0165] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0166] The systems and techniques described herein can be implemented on a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a graphical user interface or a web browser).

[0167] The system components may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0168] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0169] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0170] The above specific embodiments do not limit the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, subcombinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A multi-angle face data collection method, comprising: Obtaining a current image captured by a camera, and determining a first face orientation corresponding to a posture angle of a face in the current image; When the first face orientation is a frontal face orientation, acquiring facial feature values ​​of the face in the current image to obtain frontal face feature values; When the frontal face feature value is obtained, continue to obtain images captured by the camera, and determine a second facial orientation corresponding to the posture angle of the face in the currently obtained image, until each of the determined second facial orientations includes at least one side face orientation, and obtain facial feature values ​​of the faces in the images corresponding to the second facial orientations of the at least one side face orientation, to obtain at least one side face feature value; Determining a target facial feature value based on the frontal facial feature value and the at least one profile facial feature value, and saving the target facial feature value; The determining of the first face orientation corresponding to the posture angle of the face in the current image includes: Performing face detection on the current image to obtain face information in the current image, wherein the face information includes a target posture angle of the face; Based on the mapping relationship between each preset pitch angle threshold interval and yaw angle threshold interval and each facial orientation, the target pitch angle and target yaw angle in the target posture angle are matched with each preset pitch angle threshold interval and yaw angle threshold interval, respectively, to determine the first facial orientation corresponding to the target posture angle; the facial orientation includes: frontal face orientation and side face orientation in multiple directions; the frontal face orientation area corresponding to the frontal face orientation is determined according to the camera type, the position of the camera and the distance between the camera and the face; the side face orientation area corresponding to the side face orientation in multiple directions is determined according to the coordinates of two points on the boundary line of the frontal face orientation area determined based on the camera edge imaging situation, and the preset facial deflection angle extreme value.

2. The method according to claim 1, wherein The facial information also includes: facial feature values.

3. The method according to claim 1, further comprising: In the case that the first face orientation is not the frontal face orientation, the process returns to the step of obtaining a current image captured by a camera, and determining a first face orientation corresponding to a posture angle of the face in the current image.

4. The method according to claim 1, wherein In the case of obtaining the frontal face feature value, continuing to obtain images captured by the camera, and determining a second facial orientation corresponding to the posture angle of the face in the currently obtained image, until each of the determined second facial orientations includes a side face orientation in at least one direction, obtaining a facial feature value of the face in the image whose second facial orientation corresponds to the side face orientation in the at least one direction, and obtaining at least one side face feature value, including: When the frontal face feature value is obtained, an image captured by a camera is acquired, and a second face orientation corresponding to a posture angle of the face in the currently acquired image is determined; determining whether the second facial orientation corresponds to a profile orientation in at least one direction; If the second facial orientation does not correspond to the side facial orientation in any direction, returning to the step of acquiring an image captured by the camera and determining the second facial orientation corresponding to the posture angle of the face in the currently acquired image; In a case where the second facial orientation corresponds to a side facial orientation in at least one direction, determining whether a side facial feature value corresponding to the second facial orientation has been obtained; If no profile feature value corresponding to the second facial orientation has been obtained, obtaining facial feature values ​​of a face in a currently obtained image to obtain a profile feature value corresponding to the second facial orientation; When the side face feature value corresponding to the side face orientation corresponding to the second face orientation has been obtained, return to the execution step: obtain the image captured by the camera, and determine the second face orientation corresponding to the posture angle of the face in the currently obtained image, until the side face feature value corresponding to the side face orientation in a preset number of directions is obtained.

5. The method according to claim 4, wherein The acquiring of facial feature values ​​of a face in the currently acquired image to obtain a side face feature value corresponding to the side face orientation of the second face includes: Obtaining facial feature values ​​of the face in the currently acquired image to obtain candidate profile feature values; Calculating a similarity value between the candidate profile feature value and the frontal face feature value; If the similarity value is greater than a preset threshold, the candidate profile feature value is used as a profile feature value corresponding to the profile orientation of the second person and cached; If the similarity value is not greater than the preset threshold, the process returns to the step of acquiring an image captured by the camera and determining a second face orientation corresponding to the face posture angle in the currently acquired image.

6. The method according to any one of claims 1 to 5, wherein: The determining a target facial feature value based on the frontal facial feature value and the at least one side facial feature value, and saving the target facial feature value, includes: A weighted average of the frontal face feature value and the at least one side face feature value is determined as the target face feature value, and the target face feature value is saved.

7. A multi-angle face data acquisition device, comprising: a first image acquisition module, configured to acquire a current image captured by a camera and determine a first face orientation corresponding to a posture angle of a face in the current image; a frontal face feature value acquisition module, configured to acquire facial feature values ​​of the face in the current image when the first face orientation is a frontal face orientation, to obtain frontal face feature values; a side face feature value acquisition module, configured to, upon obtaining the frontal face feature value, continue to acquire images captured by the camera, and determine a second facial orientation corresponding to a posture angle of the face in the currently acquired image, until each of the determined second facial orientations includes a side face orientation in at least one direction, and acquire a facial feature value of the face in the image corresponding to the second facial orientation in the at least one direction, thereby obtaining at least one side face feature value; a target feature value determination module, configured to determine a target facial feature value based on the frontal facial feature value and the at least one profile facial feature value, and save the target facial feature value; The determining of the first face orientation corresponding to the posture angle of the face in the current image includes: Performing face detection on the current image to obtain face information in the current image, wherein the face information includes a target posture angle of the face; Based on the mapping relationship between each preset pitch angle threshold interval and yaw angle threshold interval and each facial orientation, the target pitch angle and target yaw angle in the target posture angle are matched with each preset pitch angle threshold interval and yaw angle threshold interval, respectively, to determine the first facial orientation corresponding to the target posture angle; the facial orientation includes: frontal face orientation and side face orientation in multiple directions; the frontal face orientation area corresponding to the frontal face orientation is determined according to the camera type, the position of the camera and the distance between the camera and the face; the side face orientation area corresponding to the side face orientation in multiple directions is determined according to the coordinates of two points on the boundary line of the frontal face orientation area determined based on the camera edge imaging situation, and the preset facial deflection angle extreme value.

8. The device according to claim 7, wherein The facial information also includes: facial feature values.

9. The apparatus according to claim 7, further comprising: The second image acquisition module is used to trigger the first image acquisition module to execute the following steps when the first face orientation is not the frontal orientation: obtaining the current image captured by the camera and determining the first face orientation corresponding to the posture angle of the face in the current image.

10. The device according to claim 7, wherein The side face feature value acquisition module includes: a first image acquisition unit, configured to acquire an image captured by a camera when the frontal face feature value is obtained, and determine a second face orientation corresponding to a posture angle of the face in the currently acquired image; a first determining unit, configured to determine whether the second facial orientation corresponds to a profile orientation in at least one direction; a second image acquisition unit configured to, when the first determination unit determines that the second facial orientation does not correspond to the side facial orientation in any direction, trigger the first image acquisition unit to execute: acquiring an image captured by the camera, and determining the second facial orientation corresponding to the posture angle of the face in the currently acquired image; a determining unit configured to determine, when the first determining unit determines that the second facial orientation corresponds to a profile orientation in at least one direction, whether a profile feature value corresponding to the profile orientation of the second facial orientation has been obtained; a side face feature value acquiring unit, configured to acquire facial feature values ​​of a face in a currently acquired image, and obtain the side face feature value in the side face orientation corresponding to the second face orientation, if the determining unit determines that no side face feature value in the side face orientation corresponding to the second face orientation has been acquired; The third image acquisition unit is used to trigger the first image acquisition unit to execute the following steps when the judgment unit determines that the side face feature value under the side face orientation corresponding to the second face orientation has been obtained: obtaining the image captured by the camera, and determining the second face orientation corresponding to the posture angle of the face in the currently obtained image, until the side face feature values ​​corresponding to the side face orientations in a preset number of directions are obtained.

11. The device according to claim 10, wherein The acquiring of facial feature values ​​of a face in the currently acquired image to obtain a side face feature value corresponding to the side face orientation of the second face includes: Obtaining facial feature values ​​of the face in the currently acquired image to obtain candidate profile feature values; Calculating a similarity value between the candidate profile feature value and the frontal face feature value; If the similarity value is greater than a preset threshold, the candidate profile feature value is used as a profile feature value corresponding to the profile orientation of the second person and cached; If the similarity value is not greater than the preset threshold, the process returns to the step of acquiring an image captured by the camera and determining a second face orientation corresponding to the face posture angle in the currently acquired image.

12. The device according to any one of claims 7 to 11, wherein: The target feature value determination module is specifically used to: A weighted average of the frontal face feature value and the at least one side face feature value is determined as the target face feature value, and the target face feature value is saved.

13. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 6.

15. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images