WO2025130660A1 - 人脸情绪识别方法、装置、存储介质及车辆 - Google Patents

人脸情绪识别方法、装置、存储介质及车辆 Download PDF

Info

Publication number
WO2025130660A1
WO2025130660A1 PCT/CN2024/137382 CN2024137382W WO2025130660A1 WO 2025130660 A1 WO2025130660 A1 WO 2025130660A1 CN 2024137382 W CN2024137382 W CN 2024137382W WO 2025130660 A1 WO2025130660 A1 WO 2025130660A1
Authority
WO
WIPO (PCT)
Prior art keywords
facial
user
features
region
closest
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/CN2024/137382
Other languages
English (en)
French (fr)
Inventor
胡束芒
林枝叶
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Great Wall Motor Co Ltd
Original Assignee
Great Wall Motor Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Great Wall Motor Co Ltd filed Critical Great Wall Motor Co Ltd
Publication of WO2025130660A1 publication Critical patent/WO2025130660A1/zh
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions

Definitions

  • the present application relates to the field of image recognition technology, and in particular to a facial emotion recognition method, device, storage medium and vehicle.
  • the embodiments of the present application provide a facial emotion recognition method, device, storage medium and vehicle to solve the technical problems in the related technology that user emotions are easily misdetected and falsely reported, and the user's emotions cannot be correctly identified.
  • an embodiment of the present application provides a method for facial emotion recognition, comprising:
  • the emotion of the user is determined.
  • extracting features from each divided facial region to obtain facial features of the user includes:
  • a facial feature matrix is obtained based on the local features and the head posture, and the facial feature matrix is used as the facial features of the user.
  • the method further includes:
  • the clustering region closest to the facial feature is obtained from the pre-established classification feature pool, including:
  • the identity of the next user is determined according to the facial image of the next user, and based on the classification feature pool associated with the identity of the next user, a clustering region closest to the facial feature of the next user is obtained.
  • the pre-established classification feature pool includes clustered regions of facial features corresponding to different emotions of different types of people;
  • the step of obtaining a classification feature pool associated with the identity of the user according to the facial features of the user and the closest clustering region includes:
  • a target clustering region is obtained from a pre-established classification feature pool; the type of personnel corresponding to the target clustering region is the same as the type of personnel corresponding to the closest clustering region;
  • a classification feature pool associated with the identity of the user is obtained.
  • obtaining a cluster region closest to the facial feature from a pre-established classification feature pool includes:
  • the clustering area corresponding to the minimum Euclidean distance among the Euclidean distances is used as the closest clustering area.
  • the method before obtaining the cluster region closest to the facial feature from the pre-established classification feature pool, the method further includes:
  • the plurality of facial images include facial images of different emotions
  • Multiple facial features of the same emotion type are clustered to obtain multiple cluster regions under the same emotion, and a classification feature pool is constructed based on all the obtained cluster regions.
  • determining the emotion of the user based on the most similar cluster area includes:
  • the emotion corresponding to the most similar cluster region is used as the emotion of the user.
  • the pre-established classification feature pool includes clustered areas of facial features corresponding to different emotions of different types of people; after acquiring the facial image of the user, the method further includes:
  • obtaining the clustering region closest to the facial feature from the pre-established classification feature pool includes:
  • a clustering area whose person type is the same as that of the user is determined in a pre-established classification feature pool, and a clustering area closest to the facial feature is determined from among the clustering areas whose person type is the same as that of the user.
  • an embodiment of the present application provides a facial emotion recognition device, comprising:
  • the acquisition module is used to acquire the user's facial image.
  • the extraction module is used to divide the facial image and extract features from each divided facial area to obtain the facial features of the user.
  • the acquisition module is used to obtain the cluster area closest to the facial feature from a pre-established classification feature pool; wherein the classification feature pool includes cluster areas of facial features corresponding to different emotions.
  • a determination module is used to determine the emotion of the user based on the most similar cluster area.
  • an embodiment of the present application provides a non-volatile computer-readable storage medium, which stores a computer program.
  • the computer program When executed by a processor, it implements the facial emotion recognition method as described in any one of the first aspects.
  • an embodiment of the present application provides a vehicle, comprising a memory and a processor, wherein the memory stores a computer program that can be executed on the processor, and when the processor executes the computer program, the facial emotion recognition method as described in any one of the first aspects is implemented.
  • the facial emotion recognition method provided in the embodiment of the present application first divides the acquired facial image of the user, extracts features from each divided facial area, and obtains the facial features of the user. Thereafter, based on a pre-established classification feature pool, a clustering area closest to the facial features is determined, wherein the classification feature pool includes clustering areas of facial features corresponding to different emotions. Furthermore, based on the above-mentioned closest clustering areas, the user's emotions are determined. In this way, the user's emotions can be determined based on the features of each area of the facial image and the closest clustering area in the classification feature pool, which can improve the accuracy of identifying the user's emotions and reduce false detections and false alarms of the user's emotions.
  • FIG1 is a flow chart of a method for facial emotion recognition provided by an embodiment of the present application.
  • FIG2 is a schematic diagram of a facial region provided by an embodiment of the present application.
  • FIG3 is a flowchart of a method for facial emotion recognition provided by an embodiment of the present application.
  • FIG4 is a flow chart of a method for recognizing facial emotions according to another embodiment of the present application.
  • FIG5 is a schematic diagram of the structure of a facial emotion recognition device provided by an embodiment of the present application.
  • FIG. 6 is a schematic diagram of the structure of a vehicle provided in an embodiment of the present application.
  • references to "one embodiment” or “some embodiments” etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements “in one embodiment”, “in some embodiments”, “in some other embodiments”, “in some other embodiments”, etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean “one or more but not all embodiments", unless otherwise specifically emphasized in other ways.
  • the terms “including”, “comprising”, “having” and their variations all mean “including but not limited to”, unless otherwise specifically emphasized in other ways.
  • the inventors have discovered through research that the user's facial image can be divided into different areas, such as the nose, mouth, forehead, and eyes, and the local features of each area can be extracted to form facial features.
  • the classification feature pool includes clustering areas of facial features corresponding to different emotions
  • the most similar clustering area is determined according to the facial features, and the user's emotions are then identified, thereby improving the accuracy of identifying the user's emotions and reducing false detections and false alarms of the user's emotions.
  • FIG1 is a flow chart of a method for facial emotion recognition provided in an embodiment of the present application. As shown in FIG1 , the method in an embodiment of the present application is applied to a vehicle, and the method may include:
  • Step 101 Obtain a user's facial image.
  • the facial image of the user may be acquired through a vehicle-mounted camera, or the facial image may be input by the user through a vehicle-mounted display screen or other terminal device.
  • Step 102 segment the facial image and extract features from each segmented facial region to obtain facial features of the user.
  • the facial image is divided into multiple facial areas, for example, the forehead, left eyebrow, right eyebrow, left eye, right eye, nose, left cheek, right cheek, mouth, chin, a total of 10 facial areas, of course, the facial image can also be divided in other ways to obtain multiple facial areas, and no specific restrictions are made here.
  • the user's head posture may also be different, for example, when happy, the head may be tilted, so in this embodiment, the head posture is also used as the basis for emotion analysis and recognition.
  • the postures of each facial area are different, for example, the mouth corresponds to relaxation, slightly open, pursed lips and upward postures, and the proportions of each posture of the mouth are different under different emotions.
  • the head posture is also different, for example, the head posture corresponds to pitch and yaw, and the values of the pitch and yaw of the head posture are different under different emotions. Therefore, the facial features include the proportions of each posture of each facial area, and the values of the pitch and yaw of the head posture.
  • this embodiment when obtaining the facial features of the user, this embodiment may perform feature extraction on each facial region, determine the local features corresponding to each facial region, and determine the head posture in the facial image of the user. Thereafter, based on the local features and the head posture, a facial feature matrix is obtained, and the facial feature matrix is used as the facial features of the user.
  • the features of each facial region can be seen in Table 1.
  • the features corresponding to each facial region are not limited to the features listed in Table 1, and can also be some other features.
  • a neural network can be used to extract features from each facial region obtained by division, and the proportions of different features corresponding to each facial region can be obtained to form a regional feature vector.
  • the features corresponding to the forehead can include relaxation, shrug and wrinkle. After local feature extraction, the proportions corresponding to the features of relaxation, shrug and wrinkle can be obtained as 97%, 0 and 3% respectively.
  • a neural network can also be used to extract features from the head posture in the facial image to obtain the head posture feature value.
  • the features corresponding to the head posture can include looking up, burying the head, tilting the head and turning the head, and the features corresponding to different head postures can be determined according to the pitch angle Pitch, yaw angle Yaw and roll angle Roll corresponding to the head.
  • the above-mentioned regional feature vector and the head posture feature value together constitute the user's facial feature matrix.
  • Step 103 Obtain the cluster region closest to the facial feature from a pre-established classification feature pool; wherein the classification feature pool includes cluster regions of facial features corresponding to different emotions.
  • the present embodiment may also collect multiple facial images, divide the multiple facial images respectively, and extract features from each facial region after the multiple facial images are divided to obtain corresponding facial features, and then cluster multiple facial features of the same emotion type to obtain multiple clustering regions under the same emotion, and construct a classification feature pool based on all the obtained clustering regions.
  • the multiple facial images include facial images of different emotions.
  • this embodiment collects facial images of people in different head postures and under different emotions.
  • multiple cameras at different angles can be used to collect facial images in different head postures.
  • this embodiment can perform region division and feature extraction on the collected multiple facial images, and perform feature extraction on the head postures in the multiple facial images to obtain facial features corresponding to each facial image, that is, to obtain a corresponding facial feature matrix.
  • the specific implementation principle and process of obtaining the facial features can refer to the relevant parts in step 102, which will not be repeated here.
  • each facial image corresponds to an emotion type
  • the above emotion types may include happiness, appreciation, sadness, anger, fear, surprise, disgust and neutrality.
  • multiple facial features of the same emotion type are clustered, and facial features with similar features in the same emotion type are divided into a cluster area.
  • the emotion corresponding to the cluster area is the above emotion type.
  • all facial features with the emotion type of happiness are clustered to obtain multiple cluster areas with the emotion type of happiness.
  • all facial features with the emotion type of sadness are clustered to obtain multiple cluster areas with the emotion type of sadness, etc.
  • all the obtained cluster areas constitute a classification feature pool.
  • each cluster area includes multiple facial features with similar features.
  • a deep learning algorithm may be used to cluster multiple facial features of the same emotion type, and facial features with similar features in the same emotion type may be divided into a clustering region, thereby constructing a classification feature pool based on all the obtained clustering regions.
  • the facial features in each cluster area in the classification feature pool with the smallest feature difference from the user's facial features are determined as the most similar facial features, and then the cluster area corresponding to the most similar facial features is determined as the most similar cluster area.
  • the most similar facial features can be determined by calculating the distance between the user's facial features and different facial features in each cluster area in the classification feature pool.
  • this embodiment can obtain the Euclidean distance between the facial features and the different clustering areas in the classification feature pool based on the vector value of the facial features and the vector value of the clustering area in the classification feature pool. Then, the clustering area corresponding to the minimum Euclidean distance in the Euclidean distance is taken as the closest clustering area.
  • the user's facial features are represented by the corresponding facial feature matrix, and the facial feature matrix is composed of regional feature vectors and head posture feature values, and each clustering area in the classification feature pool is also obtained by clustering facial features. Therefore, this embodiment can perform Euclidean distance calculation based on the vector value of the user's facial features and the vector value of different facial features in each clustering area in the classification feature pool to obtain the Euclidean distance between the user's facial features and the different facial features in each clustering area. After that, the clustering area corresponding to the minimum Euclidean distance can be determined as the clustering area closest to the user's facial features.
  • Step 104 Determine the user's emotion based on the most similar clustering area.
  • this embodiment can obtain the emotion corresponding to the closest cluster region from the pre-established classification feature pool, and use the emotion (ie, emotion type) corresponding to the closest cluster region as the emotion of the user.
  • the facial emotion recognition method provided in the embodiment of the present application first divides the acquired facial image of the user, extracts features from each divided facial area, and obtains the facial features of the user. Thereafter, based on a pre-established classification feature pool, a clustering area closest to the facial features is determined, wherein the classification feature pool includes clustering areas of facial features corresponding to different emotions. Furthermore, based on the above-mentioned closest clustering areas, the user's emotions are determined. In this way, the user's emotions can be determined based on the features of each area of the facial image and the closest clustering area in the classification feature pool, which can improve the accuracy of identifying the user's emotions and reduce false detections and false alarms of the user's emotions.
  • this embodiment can also determine the user's person type based on the user's facial image. Accordingly, when obtaining the clustering area closest to the facial features, this embodiment can determine the clustering area with the same person type as the user's person type in a pre-established classification feature pool based on the user's person type, and determine the clustering area closest to the facial features from the clustering area with the same person type as the user's person type.
  • the classification feature pool includes clustering areas of facial features corresponding to different emotions of different types of people.
  • this embodiment collects facial images of different people in different head postures and different emotions, mainly covering people of different body shapes, different ages and different appearance features.
  • the appearance features can be mainly distinguished by eyes, nose, mouth, hair, face shape and skin color, etc., covering as many appearance features related to emotions as possible.
  • the facial features corresponding to each facial image are obtained.
  • the specific implementation principle and process of obtaining the facial features can refer to the relevant parts in step 102, which will not be repeated here.
  • each facial image corresponds to an emotion type.
  • each facial image also corresponds to other types, such as age type and body type, etc.
  • the body type can include fat, moderate and thin, etc.
  • the age type can include children, teenagers, youth, middle-aged and elderly, etc.
  • a certain facial image corresponds to types such as happy, fat and children.
  • the facial features obtained according to the facial image also correspond to emotion types, age types and body types, etc.
  • clustering processing is performed on multiple facial features of the same emotion type. This is because different people, different body types or people of different age groups have different expressions of the same emotion.
  • the classification feature pool includes multiple cluster areas of different emotions, and there are also multiple cluster areas under the same emotion, that is, the same emotion is subdivided, which provides support for the subsequent accurate identification of user emotions.
  • each facial feature in each cluster area has the same emotion type, but the age type and body type of each facial feature may be different.
  • the age type and body type can be defined as the person type.
  • the person type of the cluster area can be determined according to the person types corresponding to the multiple facial features included therein, such as taking the person type with the largest number corresponding to the multiple facial features as the person type corresponding to the cluster area.
  • a cluster area includes 10 facial features, of which the age type corresponding to 9 facial features is children, and the age type corresponding to 1 facial feature is youth, then the age type of the cluster area is determined to be children.
  • the body type of the cluster area can be determined according to the body type of the facial features, etc. In this way, the type corresponding to the cluster area can be obtained, for example, the type corresponding to a cluster area is happy, fat and child, and the type corresponding to another cluster area is happy, thin and youth.
  • this embodiment can obtain the user's personnel type (i.e., age type and body type) based on the user's facial image based on the age algorithm and the body shape algorithm. Then, directly based on the user's personnel type, multiple clustering areas with the same personnel type as the user are selected from the classification feature pool. Then, based on the above-selected multiple clustering areas, the clustering area closest to the user's facial features is determined. Then, the user's emotions are determined based on the closest clustering area, which can improve the accuracy of user emotion recognition and improve recognition efficiency.
  • the user's personnel type i.e., age type and body type
  • this embodiment may also determine the user's identity based on the user's facial image, and obtain a classification feature pool associated with the user's identity based on the user's facial features and the closest clustering region.
  • the user is identified based on the user's facial image, for example, the user's facial image is identified by features to obtain an identification result, wherein the identification result may be a biometric identification result.
  • an identity such as identity A
  • identity A may be assigned to the above identification result, and a corresponding relationship between the identification result and the identity may be established based on the identification result and the corresponding identity A.
  • this embodiment may also obtain a target clustering region from a pre-established classification feature pool, and obtain a classification feature pool associated with the user's identity based on the user's facial features and the target clustering region.
  • the type of person corresponding to the target clustering region is the same as the type of person corresponding to the closest clustering region.
  • clustering regions correspond to emotion types and person types (person types include age types and body types).
  • person types include age types and body types.
  • a certain clustering region corresponds to a person type of youth and thinness
  • another clustering region corresponds to a person type of child and fatness.
  • all clustering regions with a person type of youth and thinness are selected from the pre-established classification feature pool as the target clustering region.
  • a classification feature pool is constructed based on the user's facial features and the target clustering region, and the constructed classification feature pool is associated with the user's identity A as a classification feature pool associated with the user's identity.
  • the classification feature pool associated with the user's identity is a classification feature pool of the same type as the user, and is associated with the user's own characteristics and habits.
  • the classification feature pool associated with the user's identity can be directly called to perform user emotion recognition, more accurately identify the user's emotions, and improve recognition efficiency.
  • the present embodiment can periodically update the classification feature pool associated with the user's identity based on the user's facial features and the closest clustering area. For example, after determining that the user's identity is identity A, it is determined whether the time interval between the current recognition of identity A and the last recognition of identity A is greater than a preset interval threshold. If the above time interval is greater than the preset interval threshold, the classification feature pool associated with the user's identity is updated; otherwise, the classification feature pool associated with the user's identity is not updated.
  • the classification feature pool associated with the user's identity is determined.
  • identity A is identified multiple times, for example, after identity A is identified three times, the personnel types corresponding to multiple closest clustering areas are recorded.
  • the target personnel type is determined, and then the target clustering area is determined according to the target personnel type, and then the classification feature pool associated with the user's identity is determined.
  • this embodiment constructs a classification feature pool associated with the user's own characteristics, habits, etc. based on the user's facial features and the closest clustering area, so that in the subsequent process, by identifying the user's identity, the classification feature pool associated with the user's identity can be directly called to perform user emotion recognition.
  • FIG4 is a flow chart of a facial emotion recognition method provided by another embodiment of the present application. As shown in FIG4 , the method in the embodiment of the present application may include:
  • Step 401 Obtain the facial image of the next user.
  • Step 402 segment the facial image of the next user, and extract features from each segmented facial region to obtain facial features of the next user.
  • step 401 to step 402 of this embodiment can refer to the above-mentioned embodiment and will not be repeated here.
  • Step 403 determine the identity of the next user according to the facial image of the next user, and obtain a clustering region that is closest to the facial features of the next user based on a classification feature pool associated with the identity of the next user.
  • feature recognition is performed on the facial image of the next user to obtain a recognition result, and the identity of the next user is determined based on the corresponding relationship between the recognition result and the identity.
  • a classification feature pool associated with the identity of the next user is determined, and then, a clustering region closest to the facial features of the next user is determined.
  • the specific implementation process and principle of determining the clustering region closest to the facial features of the next user based on the classification feature pool associated with the identity of the next user can refer to step 103 in the aforementioned embodiment, and will not be repeated here.
  • the clustering region closest to the facial features of the next user is determined according to step 103 in the above embodiment.
  • Step 404 Determine the emotion of the next user based on the closest cluster area.
  • step 404 of this embodiment can refer to the above-mentioned embodiment and will not be described again here.
  • This embodiment can determine the identity of the user, and directly determine the clustering area closest to the user's facial features based on the classification feature pool associated with the user's identity, and then determine the user's emotions. Since the classification feature pool associated with the user's identity is associated with the user's own characteristics and habits, the user's emotions can be more accurately identified based on the classification feature pool associated with the user's identity, reducing false detection and false alarm of user emotions, while improving recognition efficiency.
  • Fig. 5 is a schematic diagram of the structure of a facial emotion recognition device provided by an embodiment of the present application.
  • the facial emotion recognition device provided by this embodiment may include: an acquisition module 501 , an extraction module 502 , an acquisition module 503 and a determination module 504 .
  • the acquisition module 501 is used to acquire the user's facial image
  • An extraction module 502 is used to divide the facial image and extract features from each divided facial region to obtain facial features of the user;
  • the acquisition module 503 is used to obtain the cluster region closest to the facial feature from a pre-established classification feature pool; wherein the classification feature pool includes cluster regions of facial features corresponding to different emotions;
  • the determination module 504 is configured to determine the emotion of the user based on the most similar clustering region.
  • the extraction module 502 is specifically used for:
  • a facial feature matrix is obtained based on the local features and the head posture, and the facial feature matrix is used as the facial features of the user.
  • the determination module 504 is further configured to:
  • a classification feature pool associated with the identity of the user is obtained according to the facial features of the user and the most similar clustering region.
  • the obtaining module 503 is specifically used for:
  • the identity of the next user is determined according to the facial image of the next user, and based on the classification feature pool associated with the identity of the next user, a clustering region closest to the facial feature of the next user is obtained.
  • the pre-established classification feature pool includes clustered areas of facial features corresponding to different emotions of different types of people; the determination module 504 is further specifically used for:
  • a target clustering region is obtained from a pre-established classification feature pool; the type of personnel corresponding to the target clustering region is the same as the type of personnel corresponding to the closest clustering region;
  • a classification feature pool associated with the identity of the user is obtained.
  • the obtaining module 503 is specifically used for:
  • the clustering area corresponding to the minimum Euclidean distance among the Euclidean distances is used as the closest clustering area.
  • the obtaining module 503 is further specifically used for:
  • the plurality of facial images include facial images of different emotions
  • Multiple facial features of the same emotion type are clustered to obtain multiple cluster regions under the same emotion, and a classification feature pool is constructed based on all the obtained cluster regions.
  • the determination module 504 is specifically configured to:
  • the emotion corresponding to the most similar cluster region is used as the emotion of the user.
  • the pre-established classification feature pool includes clustered areas of facial features corresponding to different emotions of different types of people; the acquisition module 501 is further specifically used for:
  • the obtaining module 503 is further specifically used for:
  • a clustering area whose person type is the same as that of the user is determined in a pre-established classification feature pool, and a clustering area closest to the facial feature is determined from among the clustering areas whose person type is the same as that of the user.
  • FIG6 is a schematic diagram of the structure of a vehicle provided in an embodiment of the present application.
  • the vehicle 600 of this embodiment includes: a processor 610 and a memory 620, wherein the memory 620 stores a computer program 621 that can be run on the processor 610.
  • the processor 610 executes the computer program 621, the steps in any of the above-mentioned method embodiments are implemented, such as steps 101 to 104 shown in FIG1.
  • the processor 610 executes the computer program 621, the functions of each module/unit in the above-mentioned device embodiments are implemented, such as the functions of modules 501 to 504 shown in FIG5.
  • the computer program 621 may be divided into one or more modules/units, one or more modules/units are stored in the memory 620 and executed by the processor 610 to complete the present application.
  • the one or more modules/units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program 621 in the vehicle 600.
  • FIG. 6 is merely an example of a vehicle and does not constitute a limitation on the vehicle, and may include more or fewer components than shown in the figure, or a combination of certain components, or different components, such as input and output devices, network access devices, buses, etc.
  • the processor 610 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general-purpose processor may be a microprocessor or any conventional processor, etc.
  • the memory 620 may be an internal storage unit of the vehicle, such as a hard disk or memory of the vehicle, or an external storage device of the vehicle, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the vehicle.
  • the above-mentioned memory 620 may also include both an internal storage unit of the vehicle and an external storage device.
  • the above-mentioned memory 620 is used to store computer programs and other programs and data required by the vehicle.
  • the memory 620 may also be used to temporarily store data that has been output or is to be output.
  • the technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration.
  • the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
  • the functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units.
  • the disclosed devices/vehicles and methods can be implemented in other ways.
  • the device/vehicle embodiments described above are only schematic.
  • the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.
  • Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
  • the integrated module/unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.
  • the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program.
  • the computer program can be stored in a non-volatile computer-readable storage medium.
  • the computer program is executed by the processor, the steps of the above-mentioned method embodiments can be implemented.
  • the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form.
  • the computer-readable medium can include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Databases & Information Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Human Computer Interaction (AREA)
  • Image Analysis (AREA)

Abstract

本申请适用于图像识别技术领域，提供了一种人脸情绪识别方法、装置、存储介质及车辆。该方法包括：获取用户的面部图像；对面部图像进行划分，并对划分后的各个面部区域进行特征提取，得到用户的面部特征；从预先建立的分类特征池中，获得与面部特征最相近的聚类区域；其中，分类特征池中包括不同情绪对应的面部特征的聚类区域；基于最相近的聚类区域，确定用户的情绪。本申请可以实现基于面部图像的各个区域的特征，与分类特征池中最相近的聚类区域确定用户情绪，能够提高识别用户的情绪的准确性，减小对用户情绪错检和误报。

Description

人脸情绪识别方法、装置、存储介质及车辆
本申请要求于2023年12月21日提交的发明名称为：人脸情绪识别方法、装置、存储介质及车辆，中国专利申请号为No.CN202311776179.5的优先权。在先申请的公开内容通过整体引用并入本申请。
技术领域
本申请涉及图像识别技术领域，具体涉及一种人脸情绪识别方法、装置、存储介质及车辆。
 背景技术
随着人工智能的发展，人脸情绪的识别已经应用在越来越多的领域。例如在车辆领域，情绪的识别对于安全驾驶、车内氛围以及车辆与用户的交互体验十分重要。智能座舱未来发展趋势是要了解车内相关人员的情绪状态，从而基于相关人员的情绪提供情感化的交互和服务能力，提高安全驾驶和智能座舱的用户体验。
然而，由于不同人员的面部存在差异，相关技术中容易对用户情绪错检和误报，不能正确识别用户的情绪。
 技术问题
有鉴于此，本申请实施例提供了一种人脸情绪识别方法、装置、存储介质及车辆，以解决相关技术中容易对用户情绪错检和误报，不能正确识别用户的情绪的技术问题。
 技术解决方案
第一方面，本申请实施例提供了一种人脸情绪识别方法，包括：
获取用户的面部图像；
对所述面部图像进行划分，并对划分后的各个面部区域进行特征提取，得到所述用户的面部特征；
从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域；其中，所述分类特征池中包括不同情绪对应的面部特征的聚类区域；
基于所述最相近的聚类区域，确定所述用户的情绪。
在第一方面的一种可能的实施方式中，所述对划分后的各个面部区域进行特征提取，得到所述用户的面部特征，包括：
对划分后的各个面部区域进行特征提取，确定所述各个面部区域对应的局部特征；
确定所述用户的面部图像中的头部姿态；
基于所述局部特征和所述头部姿态，获得面部特征矩阵，将所述面部特征矩阵作为所述用户的面部特征。
在第一方面的一种可能的实施方式中，在所述基于所述最相近的聚类区域，确定所述用户的情绪之后，还包括：
根据所述用户的面部图像确定所述用户的身份；
根据所述用户的面部特征和所述最相近的聚类区域，获得与所述用户的身份关联的分类特征池；
相应的，对于下一用户，所述从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域，包括：
根据所述下一用户的面部图像确定所述下一用户的身份，并基于与所述下一用户的身份关联的分类特征池，获得与所述下一用户的面部特征最相近的聚类区域。
在第一方面的一种可能的实施方式中，所述预先建立的分类特征池中包括不同类型人员的不同情绪对应的面部特征的聚类区域；
所述根据所述用户的面部特征和所述最相近的聚类区域，获得与所述用户的身份关联的分类特征池，包括：
从预先建立的分类特征池中，获得目标聚类区域；所述目标聚类区域对应的人员类型与所述最相近的聚类区域对应的人员类型相同；
基于所述用户的面部特征和所述目标聚类区域，获得与所述用户的身份关联的分类特征池。
在第一方面的一种可能的实施方式中，所述从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域，包括：
基于所述面部特征的向量值与所述分类特征池中聚类区域的向量值，获得所述面部特征与所述分类特征池中不同聚类区域的欧式距离；
将所述欧式距离中最小欧式距离对应的聚类区域作为所述最相近的聚类区域。
在第一方面的一种可能的实施方式中，在所述从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域之前，还包括：
采集多个面部图像；所述多个面部图像包括不同情绪的面部图像；
对所述多个面部图像分别进行划分，并对所述多个面部图像划分后的各个面部区域进行特征提取，得到对应的面部特征；
对相同情绪类型的多个面部特征进行聚类，得到相同情绪下的多个聚类区域，并基于得到的所有聚类区域，构建分类特征池。
在第一方面的一种可能的实施方式中，所述基于所述最相近的聚类区域，确定所述用户的情绪，包括：
确定所述最相近的聚类区域对应的情绪；
将所述最相近的聚类区域对应的情绪作为所述用户的情绪。
在第一方面的一种可能的实施方式中，所述预先建立的分类特征池中包括不同类型人员的不同情绪对应的面部特征的聚类区域；所述获取用户的面部图像之后，还包括：
根据所述用户的面部图像确定所述用户的人员类型；
相应的，所述从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域，包括：
根据所述用户的人员类型，在预先建立的分类特征池中，确定人员类型与所述用户的人员类型相同的聚类区域，并从所述人员类型与所述用户的人员类型相同的聚类区域中，确定与所述面部特征最相近的聚类区域。
第二方面，本申请实施例提供了一种人脸情绪识别装置，包括：
获取模块，用于获取用户的面部图像。
提取模块，用于对所述面部图像进行划分，并对划分后的各个面部区域进行特征提取，得到所述用户的面部特征。
获得模块，用于从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域；其中，所述分类特征池中包括不同情绪对应的面部特征的聚类区域。
确定模块，用于基于所述最相近的聚类区域，确定所述用户的情绪。
第三方面，本申请实施例提供了一种非易失性计算机可读存储介质，计算机可读存储介质存储有计算机程序，计算机程序被处理器执行时实现如第一方面任一项所述的人脸情绪识别方法。
第四方面，本申请实施例提供了一种车辆，包括存储器和处理器，存储器中存储有可在处理器上运行的计算机程序，处理器执行计算机程序时实现如第一方面任一项所述的人脸情绪识别方法。
 有益效果
本申请实施例提供的人脸情绪识别方法，首先，对获取到的用户的面部图像进行划分，对划分后的各个面部区域进行特征提取，得到用户的面部特征，之后，基于预先建立的分类特征池，确定与面部特征最相近的聚类区域，其中，分类特征池中包括不同情绪对应的面部特征的聚类区域，进而，根据上述最相近的聚类区域，确定用户的情绪，这样，可以实现基于面部图像的各个区域的特征，与分类特征池中最相近的聚类区域确定用户情绪，能够提高识别用户的情绪的准确性，减小对用户情绪错检和误报。
可以理解的是，上述第二方面至第四方面的有益效果可以参见上述第一方面中的相关描述，在此不再赘述。
应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，并不能限制本说明书。
 附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1是本申请一实施例提供的人脸情绪识别方法的流程示意图；
图2是本申请一实施例提供的面部区域的示意图；
图3是本申请一实施例提供的人脸情绪识别方法的流程框图；
图4是本申请另一实施例提供的人脸情绪识别方法的流程示意图；
图5是本申请一实施例提供的人脸情绪识别装置的结构示意图；
图6是本申请一实施例提供的车辆的结构示意图。
 本发明的实施方式
下面结合具体实施例对本申请进行更清楚的说明。以下实施例将有助于本领域的技术人员进一步理解本申请的作用，但不以任何形式限制本申请。应当指出的是，对本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进。这些都属于本申请的保护范围。
应当理解，当在本申请说明书和所附权利要求书中使用时，术语“包括”指示所描述特征、整体、步骤、操作、元素和/或组件的存在，但并不排除一个或多个其它特征、整体、步骤、操作、元素、组件和/或其集合的存在或添加。
还应当理解，在本申请说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合，并且包括这些组合。
在本申请说明书和所附权利要求书的描述中，术语“第一”、“第二”、“第三”等仅用于区分描述，而不能理解为指示或暗示相对重要性。
在本申请说明书中描述的参考“一个实施例”或“一些实施例”等意味着在本申请的一个或多个实施例中包括结合该实施例描述的特定特征、结构或特点。由此，在本说明书中的不同之处出现的语句“在一个实施例中”、“在一些实施例中”、“在其他一些实施例中”、“在另外一些实施例中”等不是必然都参考相同的实施例，而是意味着“一个或多个但不是所有的实施例”，除非是以其他方式另外特别强调。术语“包括”、“包含”、“具有”及它们的变形都意味着“包括但不限于”，除非是以其他方式另外特别强调。
此外，本申请实施例中提到的“多个”应当被解释为两个或两个以上。
在车辆领域，智能座舱未来发展趋势是要了解车内相关人员的情绪状态，从而基于相关人员的情绪提供情感化的交互和服务能力，提高安全驾驶和智能座舱的用户体验。然而，由于不同人员的面部存在差异，相关技术中容易对用户情绪错检和误报，不能正确识别用户的情绪。
针对上述不能正确识别用户情绪的问题，发明人经研究发现，可以对用户的面部图像进行不同区域的划分，例如划分鼻子、嘴巴、额头和眼睛等区域，分别提取各个区域的局部特征构成面部特征，从而，基于预先建立的分类特征池（该分类特征池中包括不同情绪对应的面部特征的聚类区域），根据面部特征确定最相近的聚类区域，进而识别用户的情绪，提高识别用户的情绪的准确性，减小对用户情绪错检和误报。
图1是本申请一实施例提供的人脸情绪识别方法的流程示意图。如图1所示，本申请实施例中的方法，应用于车辆，该方法可以包括：
步骤101、获取用户的面部图像。
示例性的，本实施例中可以通过车载摄像头获取用户的面部图像，也可以通过车载显示屏或其他终端设备由用户输入面部图像。
步骤102、对面部图像进行划分，并对划分后的各个面部区域进行特征提取，得到用户的面部特征。
本实施例中，当用户的情绪不同时，用户面部各个区域的表情会不同，因此可以对用户面部进行区域划分，从而基于各个面部区域对应的特征进行情绪分析识别。参照图2，对面部图像划分得到多个面部区域，例如划分得到额头、左眉毛、右眉毛、左眼、右眼、鼻子、左脸颊、右脸颊、嘴巴、下巴，共计10个面部区域，当然也可以是以其他划分方式对面部图像进行划分，得到多个面部区域，在此不做具体限制。此外，当用户的情绪不同时，用户的头部姿态可能也不同，例如当开心时可能会歪头，因此本实施例中也将头部姿态作为情绪分析识别的基础。
由前述可知，当情绪不同时，各个面部区域的姿态不同，例如嘴巴对应放松、微张、抿嘴和上扬等姿态，且不同情绪下嘴巴的各个姿态对应的比例不同。类似的，当情绪不同时，头部姿态也不同，例如头部姿态对应俯仰和偏转等，不同情绪下头部姿态的俯仰和偏转等的数值不同。因此，面部特征包括各面部区域的各个姿态的比例，以及头部姿态的俯仰和偏转等的数值。对各面部区域进行特征提取，以及对头部姿态进行特征提取，可以得到用户的面部特征。
在一种可能的实施方式中，参照图3，本实施例在得到用户的面部特征时，可以对各个面部区域进行特征提取，确定各个面部区域对应的局部特征，并确定用户的面部图像中的头部姿态，之后，基于局部特征和头部姿态，获得面部特征矩阵，将面部特征矩阵作为用户的面部特征。
示例性的，各个面部区域的特征可以参见表1。其中，各个面部区域对应的特征不局限于表1中列举出的特征，还可以为其他一些特征。本实施例中，可以利用神经网络对划分得到的各个面部区域进行特征提取，得到各个面部区域对应的不同特征的比例，构成区域特征向量。例如，额头对应的特征可以包括放松、上耸和皱紧，经局部特征提取后，可以得到特征放松、上耸和皱紧对应的比例分别97%、0和3%。本实施例中还可以利用神经网络对面部图像中的头部姿态进行特征提取，以得到头部姿态特征值。例如，头部姿态对应的特征可以包括仰头、埋头、歪头和转头，而不同头部姿态对应的特征可以根据头部对应的俯仰角Pitch、偏航角Yaw和滚转角Roll确定。上述区域特征向量与头部姿态特征值共同构成用户的面部特征矩阵。
表1
面部区域 特征
额头 放松 上耸 皱紧……
眉毛 放松 上挑 下耸 挤拢……
眼睛 聚焦 涣散 呆滞 瞪大 眯眼 缩瞳 仰视 俯视 左右斜视 对眼 流泪……
鼻子 放松 外扩 内缩……
脸颊 放松 上扬 鼓起 收紧 下垮……
嘴巴 放松 微张 左右歪嘴 咬嘴唇 抿嘴 嘴角上扬 嘴角下垂 露齿 吐舌头 嘟嘴 O型嘴……
下巴 放松 上仰 下压……
头部姿态 仰头 埋头 歪头 转头……
步骤103、从预先建立的分类特征池中，获得与面部特征最相近的聚类区域；其中，上述分类特征池中包括不同情绪对应的面部特征的聚类区域。
在一种可能的实施方式中，参照图3，本实施例在获得与面部特征最相近的聚类区域之前，还可以采集多个面部图像，对多个面部图像分别进行划分，并对多个面部图像划分后的各个面部区域进行特征提取，得到对应的面部特征，之后，对相同情绪类型的多个面部特征进行聚类，得到相同情绪下的多个聚类区域，并基于得到的所有聚类区域，构建分类特征池。其中，上述多个面部图像包括不同情绪的面部图像。
示例性的，在构建分类特征池时，为保证分类特征池的全面性和多样性，本实施例采集人群在不同头部姿态下以及不同情绪下的面部图像。在采集不同面部图像时，可以利用不同角度的多个相机采集不同头部姿态下的面部图像。之后，本实施例可以对采集到的多个面部图像分别进行区域划分和特征提取，以及对多个面部图像中的头部姿态进行特征提取，得到每一个面部图像对应的面部特征，也就是得到对应的面部特征矩阵，上述得到面部特征的具体实现原理和过程可以参照步骤102中的相关部分，此处不再赘述。
可选的，本实施例中，在采集多个面部图像时，各个面部图像对应有情绪类型，上述情绪类型可以包括快乐、欣赏、悲伤、愤怒、恐惧、惊讶、厌恶和中性等。之后，对相同情绪类型的多个面部特征做聚类处理，将相同情绪类型中，特征相似的面部特征划分到一个聚类区域，该聚类区域对应的情绪就是上述情绪类型，例如，将情绪类型为快乐的所有面部特征进行聚类处理，得到多个情绪类型为快乐的聚类区域，依次类推，对情绪类型为悲伤的所有面部特征进行聚类处理，得到多个情绪类型为悲伤的聚类区域等，从而，得到的所有聚类区域构成了分类特征池。这样，每个聚类区域中包括特征相似的多个面部特征。
示例性的，也可以利用深度学习算法，对相同情绪类型的多个面部特征进行聚类处理，将相同情绪类型中，特征相似的面部特征划分到一个聚类区域，从而基于得到的所有聚类区域，构建分类特征池。
在一种可能的实施方式中，本实施例中，将分类特征池中各聚类区域中的面部特征，与用户的面部特征的特征差异最小的面部特征，确定为最相近的面部特征，进而，将最相近的面部特征对应的聚类区域，确定为最相近的聚类区域。其中，由于特征之间的距离越小特征相似度越高，因此可以通过计算用户的面部特征与分类特征池中各聚类区域中不同面部特征的距离，来确定最相近的面部特征。
例如本实施例在获得与面部特征最相近的聚类区域时，可以基于面部特征的向量值与分类特征池中聚类区域的向量值，获得面部特征与分类特征池中不同聚类区域的欧式距离，之后，将欧式距离中最小欧式距离对应的聚类区域作为最相近的聚类区域。
由前述可知，用户的面部特征由对应的面部特征矩阵表征，面部特征矩阵由区域特征向量与头部姿态特征值共同构成，而分类特征池中各聚类区域也是由面部特征聚类得到，因此，本实施例可以基于用户的面部特征的向量值，与分类特征池中各聚类区域中不同面部特征的向量值，进行欧式距离计算，得到用户的面部特征与各聚类区域中不同面部特征之间的欧式距离，之后，可以确定最小欧式距离对应的聚类区域，作为与用户的面部特征最相近的聚类区域。
步骤104、基于最相近的聚类区域，确定用户的情绪。
示例性的，参照图3，本实施例可以从预先建立的分类特征池中，获得最相近的聚类区域对应的情绪，将最相近的聚类区域对应的情绪（即情绪类型）作为用户的情绪。
本申请实施例提供的人脸情绪识别方法，首先，对获取到的用户的面部图像进行划分，对划分后的各个面部区域进行特征提取，得到用户的面部特征，之后，基于预先建立的分类特征池，确定与面部特征最相近的聚类区域，其中，分类特征池中包括不同情绪对应的面部特征的聚类区域，进而，根据上述最相近的聚类区域，确定用户的情绪，这样，可以实现基于面部图像的各个区域的特征，与分类特征池中最相近的聚类区域确定用户情绪，能够提高识别用户的情绪的准确性，减小对用户情绪错检和误报。
在一种可能的实施方式中，为进一步提高用户情绪识别的准确性和效率，本实施例在获取用户的面部图像之后，还可以根据用户的面部图像确定用户的人员类型。相应的，本实施例在获得与面部特征最相近的聚类区域时，可以根据用户的人员类型，在预先建立的分类特征池中，确定人员类型与用户的人员类型相同的聚类区域，并从人员类型与用户的人员类型相同的聚类区域中，确定与面部特征最相近的聚类区域。其中，分类特征池中包括不同类型人员的不同情绪对应的面部特征的聚类区域。
示例性的，在构建分类特征池时，为保证分类特征池的全面性和多样性，本实施例采集不同人群在不同头部姿态下以及不同情绪下的面部图像，主要覆盖不同体型、不同年龄以及不同外貌特征的人群。其中，外貌特征可以以眼睛、鼻子、嘴巴、头发、脸型和肤色等做主要区分，尽可能涵盖更多与情绪相关的外貌特征。之后，得到每一个面部图像对应的面部特征，上述得到面部特征的具体实现原理和过程可以参照步骤102中的相关部分，此处不再赘述。
由前述可知，在采集多个面部图像时，各个面部图像对应有情绪类型，本实施例中各个面部图像还对应有其他类型，例如年龄类型和体型类型等，体型类型可以包括胖、适中和瘦等，年龄类型可以包括儿童、少年、青年、中年和老年等。例如，某一面部图像对应快乐、胖和儿童等类型。则根据面部图像得到的面部特征也对应有情绪类型、年龄类型和体型类型等。之后，对相同情绪类型的多个面部特征做聚类处理，这是由于不同人、不同体型或不同年龄段的人对相同情绪有不同表现，例如对于快乐，有人表现出眯眼、抿嘴和鼻子外扩，有人表现出眯眼、张嘴和歪头，则经过聚类处理，将相似度高的面部特征聚类在一个聚类区域，形成相同情绪下的不同聚类区域，也就是说，分类特征池中包括多个不同情绪的聚类区域，同时相同情绪下也有多个聚类区域，即实现了对相同情绪进行细分，为后续准确识别用户情绪提供了支撑。
示例性的，本实施例中每一个聚类区域中的各面部特征具有相同的情绪类型，然而各面部特征的年龄类型和体型类型可能不相同。其中，可以将年龄类型和体型类型定义为人员类型。则对于每一个聚类区域，可以根据其包括的多个面部特征对应的人员类型，确定该聚类区域的人员类型，如将多个面部特征对应的数量最多的人员类型，作为该聚类区域对应的人员类型。例如，某一聚类区域包括10个面部特征，其中9个面部特征对应的年龄类型为儿童，1个面部特征对应的年龄类型为青年，则确定该聚类区域的年龄类型为儿童。类似的，可以根据面部特征的体型类型等，确定该聚类区域的体型类型。这样，可以得到聚类区域对应的类型，例如某一聚类区域对应的类型为快乐、胖和儿童，另一聚类区域对应的类型为快乐、瘦和青年。
这样，本实施例可以根据用户的面部图像，基于年龄算法和体型算法，得到用户的人员类型（即年龄类型和体型类型），之后，直接根据用户的人员类型，从分类特征池中，选取人员类型与用户的人员类型相同的多个聚类区域，进而，基于上述选取的多个聚类区域，确定与用户的面部特征最相近的聚类区域，之后，再根据最相近的聚类区域确定用户的情绪，可以提高用户情绪识别的准确性，同时提高识别效率。
为进一步提高用户情绪识别的准确性和效率，还可以基于用户的面部特征和最相近的聚类区域等，构建与用户自身特征、习惯相关联的分类特征池，从而下次对该用户进行情绪识别时，可以直接使用与用户关联的分类特征池确定该用户的情绪。
在一种可能的实施方式中，本实施例在确定用户的情绪之后，还可以根据用户的面部图像确定用户的身份，并根据用户的面部特征和最相近的聚类区域，获得与用户的身份关联的分类特征池。
示例性的，本实施例中基于用户的面部图像对用户进行身份识别，例如对用户的面部图像进行特征识别得到识别结果，其中，识别结果可以为生物特征识别结果。之后，可以赋予上述识别结果一个身份，如身份A，并根据该识别结果与相应的身份A构建识别结果与身份的对应关系。
在一种可能的实施方式中，本实施例在获得与用户的身份关联的分类特征池时，还可以从预先建立的分类特征池中，获得目标聚类区域，并基于用户的面部特征和目标聚类区域，获得与用户的身份关联的分类特征池。其中，目标聚类区域对应的人员类型与最相近的聚类区域对应的人员类型相同。
由前述实施例可知，聚类区域对应有情绪类型和人员类型（人员类型包括年龄类型和体型类型），例如某一聚类区域对应的人员类型为青年和瘦，另一聚类区域对应的人员类型为儿童和胖。本实施例中，如最相近的聚类区域对应的人员类型为青年和瘦，则从预先建立的分类特征池中，选出人员类型为青年和瘦的所有聚类区域，作为目标聚类区域。从而，根据用户的面部特征和目标聚类区域，构建分类特征池，并将构建得到的分类特征池，与该用户的身份A进行关联，作为与该用户的身份关联的分类特征池。这样，与用户的身份关联的分类特征池，是与用户的类型相同的分类特征池，与用户自身特征、习惯相关联，这样，在后续过程中，通过识别用户的身份，可以直接调用与用户的身份关联的分类特征池，进行用户情绪识别，更准确识别用户的情绪，提高识别效率。
需要说明的是，为保证与用户的身份关联的分类特征池中，各聚类区域的面部特征在与用户自身特性、习惯相关联的前提下，尽可能地保留全面性和多样性，本实施例可以周期性地根据用户的面部特征和最相近的聚类区域，更新与用户的身份关联的分类特征池，例如，在确定用户的身份为身份A后，判断本次识别到身份A与上次识别到身份A的时间间隔是否大于预设间隔阈值，若上述时间间隔大于预设间隔阈值，则更新与用户的身份关联的分类特征池，否则，不更新与用户的身份关联的分类特征池。
需要注意的是，为提高与用户的身份关联的分类特征池的关联性，也可以在确定用户的身份为身份A后，仅记录最相近的聚类区域对应的人员类型，并不构建与用户的身份关联的分类特征池，而是在多次识别到身份A后，例如3次识别到身份A后，记录有多个最相近的聚类区域对应的人员类型，从而，根据上述多个最相近的聚类区域对应的人员类型，确定目标人员类型，进而根据目标人员类型确定目标聚类区域，之后，再确定与用户的身份关联的分类特征池。
这样，本实施例根据用户的面部特征和最相近的聚类区域，构建与用户自身特征、习惯等相关联的分类特征池，从而在后续过程中，通过识别用户的身份，可以直接调用与用户的身份关联的分类特征池，进行用户情绪识别。
图4是本申请另一实施例提供的人脸情绪识别方法的流程示意图。如图4所示，本申请实施例中的方法，可以包括：
步骤401、获取下一用户的面部图像。
步骤402、对下一用户的面部图像进行划分，并对划分后的各个面部区域进行特征提取，得到下一用户的面部特征。
本实施例步骤401至步骤402的具体实现过程和原理可以参照前述实施例，此处不再赘述。
步骤403、根据下一用户的面部图像确定下一用户的身份，并基于与下一用户的身份关联的分类特征池，获得与下一用户的面部特征最相近的聚类区域。
示例性的，本实施例中，对下一用户的面部图像进行特征识别得到识别结果，并根据识别结果与身份的对应关系，确定下一用户的身份。从而，根据下一用户的身份，确定与下一用户的身份关联的分类特征池，进而，确定与下一用户的面部特征最相近的聚类区域。其中，根据下一用户的身份关联的分类特征池，确定与下一用户的面部特征最相近的聚类区域的具体实现过程和原理可以参照前述实施例中步骤103，此处不再赘述。
需要说明的是，若没有与下一用户的身份关联的分类特征池，即下一用户为新用户，则根据前述实施例中步骤103确定与下一用户的面部特征最相近的聚类区域。
步骤404、基于最相近的聚类区域，确定下一用户的情绪。
本实施例步骤404的具体实现过程和原理可以参照前述实施例，此处不再赘述。
本实施例可以确定用户的身份，并直接根据与用户的身份关联的分类特征池，确定与用户的面部特征最相近的聚类区域，进而确定用户的情绪，由于与用户的身份关联的分类特征池与用户自身特征、习惯相关联，因此基于与用户的身份关联的分类特征池，可以更准确识别用户的情绪，减小对用户情绪错检和误报，同时提高识别效率。
应理解，上述实施例中各步骤的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
图5是本申请一实施例提供的人脸情绪识别装置的结构示意图。如图5所示，本实施例提供的人脸情绪识别装置，可以包括：获取模块501、提取模块502、获得模块503和确定模块504。
其中，获取模块501，用于获取用户的面部图像；
提取模块502，用于对所述面部图像进行划分，并对划分后的各个面部区域进行特征提取，得到所述用户的面部特征；
获得模块503，用于从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域；其中，所述分类特征池中包括不同情绪对应的面部特征的聚类区域；
确定模块504，用于基于所述最相近的聚类区域，确定所述用户的情绪。
可选的，提取模块502具体用于：
对各个面部区域进行特征提取，确定所述各个面部区域对应的局部特征；
确定所述用户的面部图像中的头部姿态；
基于所述局部特征和所述头部姿态，获得面部特征矩阵，将所述面部特征矩阵作为所述用户的面部特征。
可选的，确定模块504还用于：
根据所述用户的面部图像确定所述用户的身份；
根据所述用户的面部特征和所述最相近的聚类区域，获得与所述用户的身份关联的分类特征池。
相应的，对于下一用户，获得模块503具体用于：
根据所述下一用户的面部图像确定所述下一用户的身份，并基于与所述下一用户的身份关联的分类特征池，获得与所述下一用户的面部特征最相近的聚类区域。
可选的，所述预先建立的分类特征池中包括不同类型人员的不同情绪对应的面部特征的聚类区域；确定模块504还具体用于：
从预先建立的分类特征池中，获得目标聚类区域；所述目标聚类区域对应的人员类型与所述最相近的聚类区域对应的人员类型相同；
基于所述用户的面部特征和所述目标聚类区域，获得与所述用户的身份关联的分类特征池。
可选的，获得模块503具体用于：
基于所述面部特征的向量值与所述分类特征池中聚类区域的向量值，获得所述面部特征与所述分类特征池中不同聚类区域的欧式距离；
将所述欧式距离中最小欧式距离对应的聚类区域作为所述最相近的聚类区域。
可选的，获得模块503还具体用于：
采集多个面部图像；所述多个面部图像包括不同情绪的面部图像；
对所述多个面部图像分别进行划分，并对所述多个面部图像划分后的各个面部区域进行特征提取，得到对应的面部特征；
对相同情绪类型的多个面部特征进行聚类，得到相同情绪下的多个聚类区域，并基于得到的所有聚类区域，构建分类特征池。
可选的，确定模块504具体用于：
确定所述最相近的聚类区域对应的情绪；
将所述最相近的聚类区域对应的情绪作为所述用户的情绪。
可选的，所述预先建立的分类特征池中包括不同类型人员的不同情绪对应的面部特征的聚类区域；获取模块501还具体用于：
根据所述用户的面部图像确定所述用户的人员类型；
相应的，获得模块503还具体用于：
根据所述用户的人员类型，在预先建立的分类特征池中，确定人员类型与所述用户的人员类型相同的聚类区域，并从所述人员类型与所述用户的人员类型相同的聚类区域中，确定与所述面部特征最相近的聚类区域。
需要说明的是，上述装置/单元之间的信息交互、执行过程等内容，由于与本申请方法实施例基于同一构思，其具体功能及带来的技术效果，具体可参见方法实施例部分，此处不再赘述。
图6是本申请一实施例提供的车辆的结构示意图。如图6所示，该实施例的车辆600包括：处理器610、存储器620，上述存储器620中存储有可在处理器610上运行的计算机程序621。处理器610执行计算机程序621时实现上述任意各个方法实施例中的步骤，例如图1所示的步骤101至104。或者，处理器610执行计算机程序621时实现上述各装置实施例中各模块/单元的功能，例如图5所示模块501至504的功能。
示例性的，计算机程序621可以被分割成一个或多个模块/单元，一个或者多个模块/单元被存储在存储器620中，并由处理器610执行，以完成本申请。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述计算机程序621在车辆600中的执行过程。
本领域技术人员可以理解，图6仅仅是车辆的示例，并不构成对车辆的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如输入输出设备、网络接入设备、总线等。
处理器610可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器 (Digital Signal Processor，DSP)、专用集成电路 (Application Specific Integrated Circuit，ASIC)、现成可编程门阵列 (Field-Programmable Gate Array，FPGA) 或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
存储器620可以是车辆的内部存储单元，例如车辆的硬盘或内存，也可以是车辆的外部存储设备，例如车辆上配备的插接式硬盘，智能存储卡（Smart Media Card，SMC），安全数字（Secure Digital，SD）卡，闪存卡（Flash Card）等。上述存储器620还可以既包括车辆的内部存储单元也包括外部存储设备。上述存储器620用于存储计算机程序以及车辆所需的其他程序和数据。存储器620还可以用于暂时地存储已经输出或者将要输出的数据。
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，仅以上述各功能单元、模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能单元、模块完成，即将所述装置的内部结构划分成不同的功能单元或模块，以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中，上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。另外，各功能单元、模块的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。上述系统中单元、模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述或记载的部分，可以参见其它实施例的相关描述。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。
在本发明所提供的实施例中，应该理解到，所揭露的装置/车辆和方法，可以通过其它的方式实现。例如，以上所描述的装置/车辆实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口，装置或单元的间接耦合或通讯连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明实现上述实施例方法中的全部或部分流程，也可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一非易失性计算机可读存储介质中，该计算机程序在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器（ROM，Read-Only Memory）、随机存取存储器（RAM，Random Access Memory）、电载波信号、电信信号以及软件分发介质等。
以上所述实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围，均应包含在本发明的保护范围之内。

Claims (15)

  1. 一种人脸情绪识别方法，其特征在于，包括：
    获取用户的面部图像；
    对所述面部图像进行划分，并对划分后的各个面部区域进行特征提取，得到所述用户的面部特征；
    从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域；其中，所述分类特征池中包括不同情绪对应的面部特征的聚类区域；
    基于所述最相近的聚类区域，确定所述用户的情绪。
  2. 根据权利要求1所述的人脸情绪识别方法，其特征在于，所述对划分后的各个面部区域进行特征提取，得到所述用户的面部特征，包括：
    对划分后的各个面部区域进行特征提取，确定所述各个面部区域对应的局部特征；
    确定所述用户的面部图像中的头部姿态；
    基于所述局部特征和所述头部姿态，获得面部特征矩阵，将所述面部特征矩阵作为所述用户的面部特征。
  3. 根据权利要求2所述的人脸情绪识别方法，其特征在于，所述对划分后的各个面部区域进行特征提取，包括：利用神经网络对划分得到的各个面部区域进行特征提取，得到各个面部区域对应的不同特征的比例，构成区域特征向量；
    所述确定所述用户的面部图像中的头部姿态，包括：利用神经网络对面部图像中的头部姿态进行特征提取，以得到头部姿态特征值；
    其中，所述区域特征向量和所述头部姿态特征值共同构成所述用户的面部特征矩阵。
  4. 根据权利要求1所述的人脸情绪识别方法，其特征在于，在所述基于所述最相近的聚类区域，确定所述用户的情绪之后，还包括：
    根据所述用户的面部图像确定所述用户的身份；
    根据所述用户的面部特征和所述最相近的聚类区域，获得与所述用户的身份关联的分类特征池；
    相应的，对于下一用户，所述从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域，包括：
    根据所述下一用户的面部图像确定所述下一用户的身份，并基于与所述下一用户的身份关联的分类特征池，获得与所述下一用户的面部特征最相近的聚类区域。
  5. 根据权利要求4所述的人脸情绪识别方法，其特征在于，所述预先建立的分类特征池中包括不同类型人员的不同情绪对应的面部特征的聚类区域；
    所述根据所述用户的面部特征和所述最相近的聚类区域，获得与所述用户的身份关联的分类特征池，包括：
    从预先建立的分类特征池中，获得目标聚类区域；所述目标聚类区域对应的人员类型与所述最相近的聚类区域对应的人员类型相同；
    基于所述用户的面部特征和所述目标聚类区域，获得与所述用户的身份关联的分类特征池。
  6. 根据权利要求5所述的人脸情绪识别方法，其特征在于，所以根据所述用户的面部特征和所述最相近的聚类区域，获得与所述用户的身份关联的分类特征池，还包括：
    根据所述用户的面部特征和所述最相近的聚类区域，周期性地更新与所述用户的身份关联的分类特征池；
    在确定所述用户的身份后，判断本次识别到身份与上次识别到同一身份的时间间隔是否大于预设间隔阈值，若所述时间间隔大于所述预设间隔阈值，则更新与所述用户的身份关联的分类特征池，否则，不更新与所述用户的身份关联的分类特征池。
  7. 根据权利要求1所述的人脸情绪识别方法，其特征在于，所述从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域，包括：
    将所述分类特征池中各聚类区域中的面部特征，与所述用户的面部特征差异最小的面部特征，确定为最相近的面部特征；
    将所述最接近的面部特征对应的聚类区域，确定为所述最相近的聚类区域。
  8. 根据权利要求1所述的人脸情绪识别方法，其特征在于，所述从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域，包括：
    基于所述面部特征的向量值与所述分类特征池中聚类区域的向量值，获得所述面部特征与所述分类特征池中不同聚类区域的欧式距离；
    将所述欧式距离中最小欧式距离对应的聚类区域作为所述最相近的聚类区域。
  9. 根据权利要求1至8任一项所述的人脸情绪识别方法，其特征在于，在所述从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域之前，还包括：
    采集多个面部图像；所述多个面部图像包括不同情绪的面部图像；
    对所述多个面部图像分别进行划分，并对所述多个面部图像划分后的各个面部区域进行特征提取，得到对应的面部特征；
    对相同情绪类型的多个面部特征进行聚类，得到相同情绪下的多个聚类区域，并基于得到的所有聚类区域，构建分类特征池。
  10. 根据权利要求9所述的人脸情绪识别方法，其特征在于，所述采集多个面部图像，包括：采集人群在不同头部姿态下以及不同情绪下的面部图像。
  11. 根据权利要求1至8任一项所述的人脸情绪识别方法，其特征在于，所述基于所述最相近的聚类区域，确定所述用户的情绪，包括：
    确定所述最相近的聚类区域对应的情绪；
    将所述最相近的聚类区域对应的情绪作为所述用户的情绪。
  12. 根据权利要求1所述的人脸情绪识别方法，其特征在于，所述预先建立的分类特征池中包括不同类型人员的不同情绪对应的面部特征的聚类区域；所述获取用户的面部图像之后，还包括：
    根据所述用户的面部图像确定所述用户的人员类型；
    相应的，所述从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域，包括：
    根据所述用户的人员类型，在预先建立的分类特征池中，确定人员类型与所述用户的人员类型相同的聚类区域，并从所述人员类型与所述用户的人员类型相同的聚类区域中，确定与所述面部特征最相近的聚类区域。
  13. 一种人脸情绪识别装置，其特征在于，包括：
    获取模块，用于获取用户的面部图像；
    提取模块，用于对所述面部图像进行划分，并对划分后的各个面部区域进行特征提取，得到所述用户的面部特征；
    获得模块，用于从预先建立的分类特征池中，获得与所述面部特征最相近的聚类区域；其中，所述分类特征池中包括不同情绪对应的面部特征的聚类区域；
    确定模块，用于基于所述最相近的聚类区域，确定所述用户的情绪。
  14. 一种非易失性计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1至12任一项所述的人脸情绪识别方法。
  15. 一种车辆，包括存储器和处理器，所述存储器中存储有可在所述处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现如权利要求1至12任一项所述的人脸情绪识别方法。
PCT/CN2024/137382 2023-12-21 2024-12-06 人脸情绪识别方法、装置、存储介质及车辆 Pending WO2025130660A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202311776179.5 2023-12-21
CN202311776179.5A CN117746483A (zh) 2023-12-21 2023-12-21 人脸情绪识别方法、装置、存储介质及车辆

Publications (1)

Publication Number Publication Date
WO2025130660A1 true WO2025130660A1 (zh) 2025-06-26

Family

ID=90281086

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2024/137382 Pending WO2025130660A1 (zh) 2023-12-21 2024-12-06 人脸情绪识别方法、装置、存储介质及车辆

Country Status (2)

Country Link
CN (1) CN117746483A (zh)
WO (1) WO2025130660A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117746483A (zh) * 2023-12-21 2024-03-22 长城汽车股份有限公司 人脸情绪识别方法、装置、存储介质及车辆

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104318221A (zh) * 2014-11-05 2015-01-28 中南大学 一种基于elm的人脸表情识别方法
CN110147729A (zh) * 2019-04-16 2019-08-20 深圳壹账通智能科技有限公司 用户情绪识别方法、装置、计算机设备及存储介质
CN115937949A (zh) * 2022-12-14 2023-04-07 上海车联天下信息技术有限公司 一种表情识别的方法、装置、电子设备及存储介质
CN117746483A (zh) * 2023-12-21 2024-03-22 长城汽车股份有限公司 人脸情绪识别方法、装置、存储介质及车辆

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104318221A (zh) * 2014-11-05 2015-01-28 中南大学 一种基于elm的人脸表情识别方法
CN110147729A (zh) * 2019-04-16 2019-08-20 深圳壹账通智能科技有限公司 用户情绪识别方法、装置、计算机设备及存储介质
CN115937949A (zh) * 2022-12-14 2023-04-07 上海车联天下信息技术有限公司 一种表情识别的方法、装置、电子设备及存储介质
CN117746483A (zh) * 2023-12-21 2024-03-22 长城汽车股份有限公司 人脸情绪识别方法、装置、存储介质及车辆

Also Published As

Publication number Publication date
CN117746483A (zh) 2024-03-22

Similar Documents

Publication Publication Date Title
Dibeklioğlu et al. Recognition of genuine smiles
Kumar et al. Attribute and simile classifiers for face verification
Kumar et al. Real time face recognition using adaboost improved fast PCA algorithm
Azzopardi et al. Fusion of domain-specific and trainable features for gender recognition from face images
KR20120069922A (ko) 얼굴 인식 장치 및 그 방법
Zafeiriou et al. Probabilistic slow features for behavior analysis
WO2025130660A1 (zh) 人脸情绪识别方法、装置、存储介质及车辆
WO2024000233A1 (zh) 面部表情识别方法、装置、设备及可读存储介质
Pallavi et al. Deep learning based application in detecting wrinkle and predicting age
Barroso et al. Periocular recognition: how much facial expressions affect performance?
Boodoo et al. Robust multi biometric recognition using face and ear images
CN115410245A (zh) 基于双目的活体检测方法、装置及存储介质
Wu et al. Spontaneous versus posed smile recognition via region-specific texture descriptor and geometric facial dynamics
Dubey et al. Deep Perusal of Human Face Recognition Algorithms from Facial Snapshots
CN113792572A (zh) 一种基于局部表征的面部表情识别方法
Khan Detection of emotions from video in non-controlled environment
Wu et al. How do you smile? Towards a comprehensive smile analysis system
Kumar et al. Emotion recognition using anatomical information in facial expressions
KR20240025797A (ko) 피부 진단 장치, 이를 포함하는 피부 진단 시스템 및 그 방법
Belle Detection and recognition of human faces using random forests for a mobile robot
Hsu et al. Extraction of visual facial features for health management
Jain et al. Face recognition
CN118781414B (zh) 一种高度近视牵拉性病变分类方法、系统、设备及介质
WO2020052525A1 (zh) 人脸面部光泽度分类装置、方法及计算机存储介质
George Exploring methodologies in detecting facial macro and micro-expressions for deception detection

Legal Events

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

Ref document number: 24906144

Country of ref document: EP

Kind code of ref document: A1