Face living body detection method and apparatus, device, and storage medium
This application provides a face living body detection method performed by a computing device, the method including: obtaining a first face image of a target detection object in a first illumination condition and a second face image of the target detection object in a second illumination condition, determining a difference image according to the two images, decoupling an object reflectivity and an object normal vector corresponding to the target detection object from a feature map extracted from the difference image, and determining whether the target detection object is a living body according to the object reflectivity and the object normal vector. This method decouples texture information and depth information of a face, and performs living body detection by using decoupled information, which increases the defense capability against 3D attacks, thereby effectively defending against planar attacks and 3D attacks.
1. A face living body detection method performed by a computing device, the method comprising:
obtaining a first face image of a target detection object in a first illumination condition and a second face image of the target detection object in a second illumination condition;
determining a difference image according to the first face image and the second face image;
extracting a feature map from the difference image, and decoupling an object reflectivity corresponding to the target detection object and an object normal vector corresponding to the target detection object from the feature map, the object reflectivity being used for representing texture information, and the object normal vector being used for representing depth information; and
determining whether the target detection object is a living body according to the object reflectivity and the object normal vector.
2. The method according to claim 1 , wherein the determining a difference image according to the first face image and the second face image comprises:
identifying a face central region of the first face image to obtain a first face central region image, and identifying a face central region of the second face image to obtain a second face central region image; and
performing image difference processing on the first face central region image and the second face central region image to obtain the difference image between the face central regions.
3. The method according to claim 1 , wherein the determining a difference image according to the first face image and the second face image comprises:
identifying a face local organ region of the first face image to obtain a first face local organ region image, and identifying a face local organ region of the second face image to obtain a second face local organ region image; and
performing image difference processing on the first face local organ region image and the second face local organ region image to obtain the difference image between the face local organ regions.
4. The method according to claim 1 , wherein the determining a difference image according to the first face image and the second face image comprises:
identifying a face central region of the first face image to obtain a first face central region image, and identifying a face central region of the second face image to obtain a second face central region image;
performing image difference processing on the first face central region image and the second face central region image to obtain a first difference image between the face central regions;
identifying a face local organ region of the first face image to obtain a first face local organ region image, and identifying a face local organ region of the second face image to obtain a second face local organ region image; and
performing image difference processing on the first face local organ region image and the second face local organ region image to obtain a second difference image between the face local organ regions.
5. The method according to claim 1 , wherein the extracting a feature map from the difference image, and decoupling an object reflectivity corresponding to the target detection object and an object normal vector corresponding to the target detection object from the feature map; and determining whether the target detection object is a living body according to the object reflectivity and the object normal vector comprises:
extracting a feature map from the difference image by using a pre-trained face living body detection model, and decoupling the object reflectivity corresponding to the target detection object and the object normal vector corresponding to the target detection object from the feature map; and determining, by using the face living body detection model, whether the target detection object is a living body according to the object reflectivity and the object normal vector.
6. The method according to claim 5 , wherein the extracting a feature map from the difference image by using a pre-trained face living body detection model, and decoupling the object reflectivity corresponding to the target detection object and the object normal vector corresponding to the target detection object from the feature map; and determining, by using the face living body detection model, whether the target detection object is a living body according to the object reflectivity and the object normal vector comprises:
inputting the difference image into the face living body detection model, obtaining the feature map by extracting an image feature by using a convolution layer in the face living body detection model, decoupling the object reflectivity corresponding to the target detection object and the object normal vector corresponding to the target detection object from the feature map, and determining, by using a global pooling layer and a fully connected classification layer in the face living body detection model, whether the target detection object is a living body according to the object reflectivity and the object normal vector.
7. The method according to claim 1 , wherein the obtaining a first face image of a target detection object in a first illumination condition and a second face image of the target detection object in a second illumination condition comprises:
generating an illumination instruction in response to a living body detection request, the illumination instruction comprising a first illumination parameter corresponding to the first illumination condition and a second illumination parameter corresponding to the second illumination condition; and
controlling a light emitting component to emit light and acquiring facial reflected light of the target detection object based on the illumination instruction to sequentially form the first face image and the second face image.
8. The method according to claim 7 , wherein the generating an illumination instruction in response to a living body detection request comprises:
randomly selecting two different color identifiers in response to the living body detection request to generate an illumination instruction carrying the two different color identifiers.
9. The method according to claim 7 , wherein the controlling a light emitting component to emit light and acquiring facial reflected light of the target detection object based on the illumination instruction to sequentially form the first face image and the second face image comprises:
determining an angle of a face relative to an image acquisition region in response to the living body detection request;
controlling, according to the illumination instruction and the angle, the light emitting component to emit light of the angle sequentially based on the first illumination parameter and the second illumination parameter; and
acquiring the facial reflected light of the target detection object to sequentially form the first face image and the second face image.
10. A computing device, comprising:
a processor and a memory,
the memory being configured to store a plurality of computer programs, and
the processor being configured to, when executing the plurality of computer programs, perform a plurality of operations including:
obtaining a first face image of a target detection object in a first illumination condition and a second face image of the target detection object in a second illumination condition;
determining a difference image according to the first face image and the second face image;
extracting a feature map from the difference image, and decoupling an object reflectivity corresponding to the target detection object and an object normal vector corresponding to the target detection object from the feature map, the object reflectivity being used for representing texture information, and the object normal vector being used for representing depth information; and
determining whether the target detection object is a living body according to the object reflectivity and the object normal vector.
11. The computing device according to claim 10 , wherein the determining a difference image according to the first face image and the second face image comprises:
identifying a face central region of the first face image to obtain a first face central region image, and identifying a face central region of the second face image to obtain a second face central region image; and
performing image difference processing on the first face central region image and the second face central region image to obtain the difference image between the face central regions.
12. The computing device according to claim 10 , wherein the determining a difference image according to the first face image and the second face image comprises:
identifying a face local organ region of the first face image to obtain a first face local organ region image, and identifying a face local organ region of the second face image to obtain a second face local organ region image; and
performing image difference processing on the first face local organ region image and the second face local organ region image to obtain the difference image between the face local organ regions.
13. The computing device according to claim 10 , wherein the determining a difference image according to the first face image and the second face image comprises:
identifying a face central region of the first face image to obtain a first face central region image, and identifying a face central region of the second face image to obtain a second face central region image;
performing image difference processing on the first face central region image and the second face central region image to obtain a first difference image between the face central regions;
identifying a face local organ region of the first face image to obtain a first face local organ region image, and identifying a face local organ region of the second face image to obtain a second face local organ region image; and
performing image difference processing on the first face local organ region image and the second face local organ region image to obtain a second difference image between the face local organ regions.
14. The computing device according to claim 10 , wherein the extracting a feature map from the difference image, and decoupling an object reflectivity corresponding to the target detection object and an object normal vector corresponding to the target detection object from the feature map; and determining whether the target detection object is a living body according to the object reflectivity and the object normal vector comprises:
extracting a feature map from the difference image by using a pre-trained face living body detection model, and decoupling the object reflectivity corresponding to the target detection object and the object normal vector corresponding to the target detection object from the feature map; and determining, by using the face living body detection model, whether the target detection object is a living body according to the object reflectivity and the object normal vector.
15. The computing device according to claim 14 , wherein the extracting a feature map from the difference image by using a pre-trained face living body detection model, and decoupling the object reflectivity corresponding to the target detection object and the object normal vector corresponding to the target detection object from the feature map; and determining, by using the face living body detection model, whether the target detection object is a living body according to the object reflectivity and the object normal vector comprises:
inputting the difference image into the face living body detection model, obtaining the feature map by extracting an image feature by using a convolution layer in the face living body detection model, decoupling the object reflectivity corresponding to the target detection object and the object normal vector corresponding to the target detection object from the feature map, and determining, by using a global pooling layer and a fully connected classification layer in the face living body detection model, whether the target detection object is a living body according to the object reflectivity and the object normal vector.
16. The computing device according to claim 10 , wherein the obtaining a first face image of a target detection object in a first illumination condition and a second face image of the target detection object in a second illumination condition comprises:
generating an illumination instruction in response to a living body detection request, the illumination instruction comprising a first illumination parameter corresponding to the first illumination condition and a second illumination parameter corresponding to the second illumination condition; and
controlling a light emitting component to emit light and acquiring facial reflected light of the target detection object based on the illumination instruction to sequentially form the first face image and the second face image.
17. The computing device according to claim 16 , wherein the generating an illumination instruction in response to a living body detection request comprises:
randomly selecting two different color identifiers in response to the living body detection request to generate an illumination instruction carrying the two different color identifiers.
18. The computing device according to claim 16 , wherein the controlling a light emitting component to emit light and acquiring facial reflected light of the target detection object based on the illumination instruction to sequentially form the first face image and the second face image comprises:
determining an angle of a face relative to an image acquisition region in response to the living body detection request;
controlling, according to the illumination instruction and the angle, the light emitting component to emit light of the angle sequentially based on the first illumination parameter and the second illumination parameter; and
acquiring the facial reflected light of the target detection object to sequentially form the first face image and the second face image.
19. A non-transitory computer-readable storage medium, configured to store a plurality of computer programs, the computer programs, when executed by a processor of a computing device, causing the computing device to perform a plurality of operations including:
obtaining a first face image of a target detection object in a first illumination condition and a second face image of the target detection object in a second illumination condition;
determining a difference image according to the first face image and the second face image;
extracting a feature map from the difference image, and decoupling an object reflectivity corresponding to the target detection object and an object normal vector corresponding to the target detection object from the feature map, the object reflectivity being used for representing texture information, and the object normal vector being used for representing depth information; and
determining whether the target detection object is a living body according to the object reflectivity and the object normal vector.
20. The non-transitory computer-readable storage medium according to claim 19 , wherein the obtaining a first face image of a target detection object in a first illumination condition and a second face image of the target detection object in a second illumination condition comprises:
generating an illumination instruction in response to a living body detection request, the illumination instruction comprising a first illumination parameter corresponding to the first illumination condition and a second illumination parameter corresponding to the second illumination condition; and
controlling a light emitting component to emit light and acquiring facial reflected light of the target detection object based on the illumination instruction to sequentially form the first face image and the second face image.