Image processing method and apparatus, computer device, and storage medium

1. A method for determining a target image region of a target object in a target image, the method comprising:

obtaining, by a device comprising a memory storing instructions and a processor in communication with the memory, a target image comprising a target object;

obtaining, by the device, an original mask and an image segmentation model, the image segmentation model comprising a first unit model and a second unit model;

downsampling, by the device, the original mask based on a pooling layer in the first unit model to obtain a downsampled mask;

extracting, by the device, region convolution feature information of the target image based on a convolution pooling layer in the second unit model and the downsampled mask;

updating, by the device, the original mask according to the region convolution feature information; and

in response to the updated original mask satisfying an error convergence condition, determining, by the device, a target image region of the target object in the target image according to the updated original mask.

2. The method according to claim 1 , wherein:

the convolution pooling layer in the second unit model comprises a first convolution pooling layer and a second convolution pooling layer;

the downsampled mask comprises a first original mask corresponding to the first convolution pooling layer and a second original mask corresponding to the second convolution pooling layer; and

the extracting the region convolution feature information of the target image based on the convolution pooling layer in the second unit model and the downsampled mask comprises:

performing convolution and pooling on the target image based on the first convolution pooling layer and the first original mask to obtain first convolution feature information, and

performing convolution and pooling on the first convolution feature information based on the second convolution pooling layer and the second original mask to obtain the region convolution feature information of the target image.

3. The method according to claim 2 , wherein:

the first convolution pooling layer comprises a first convolution pooling unit and a second convolution pooling unit; and

the performing convolution and pooling on the target image based on the first convolution pooling layer and the first original mask to obtain the first convolution feature information comprises:

performing convolution and pooling on the target image based on a convolution function and a pooling function corresponding to the first convolution pooling unit to obtain input convolution feature information,

encoding the input convolution feature information based on a convolution function corresponding to the second convolution pooling unit to generate second convolution feature information,

performing pooling on feature maps of a plurality of channels of the second convolution feature information according to the first original mask separately to determine a pooling vector,

converting the pooling vector into a target vector based on an activation function corresponding to the second convolution pooling unit, and

generating the first convolution feature information according to the target vector and the first original mask, the input convolution feature information, the first convolution feature information, and the second convolution feature information having a same size.

4. The method according to claim 2 , wherein the updating the original mask according to the region convolution feature information comprises:

performing deconvolution on the region convolution feature information to generate third convolution feature information;

superimposing the third convolution feature information and the first convolution feature information into target convolution feature information; and

performing deconvolution and full connection on the target convolution feature information to obtain a to-be-updated mask, and determining the to-be-updated mask as the original mask, the to-be-updated mask and the target image having a same size.

5. The method according to claim 4 , wherein:

the to-be-updated mask comprises a plurality of to-be-updated mask-units; and

the determining the to-be-updated mask as the original mask comprises:

in response to a value of a to-be-updated mask-unit in the to-be-updated mask being less than or equal to a mask threshold, adjusting the value of the to-be-updated mask-unit to a first numerical value, and

in response to the value of the to-be-updated mask-unit in the to-be-updated mask being greater than the mask threshold, adjusting the value of the to-be-updated mask-unit to a second numerical value, to obtain the original mask.

6. The method according to claim 1 , wherein the determining the target image region of the target object in the target image according to the updated original mask comprises:

determining the updated original mask as a target mask, the target mask comprising a plurality of target mask-units;

determining location information of a target mask-unit greater than a mask threshold in the target mask; and

using an image region corresponding to the location information in the target image as the target image region.

7. The method according to claim 1 , further comprising:

obtaining a sample image comprising the target object, and obtaining a sample original mask and a sample image segmentation model, the sample image segmentation model comprising the first unit model and a sample second unit model;

downsampling the sample original mask based on the pooling layer in the first unit model to obtain a sample downsampled mask;

generating a prediction mask based on the sample second unit model and the sample downsampled mask;

determining a sample mask according to an image region of the target object in the sample image; and

training the sample second unit model according to the sample image, the prediction mask, and the sample mask to obtain the image segmentation model.

8. An apparatus for determining a target image region of a target object in a target image, the apparatus comprising:

a memory storing instructions; and

a processor in communication with the memory, wherein, when the processor executes the instructions, the processor is configured to cause the apparatus to perform:

obtaining a target image comprising a target object,

obtaining an original mask and an image segmentation model, the image segmentation model comprising a first unit model and a second unit model,

downsampling the original mask based on a pooling layer in the first unit model to obtain a downsampled mask,

extracting region convolution feature information of the target image based on a convolution pooling layer in the second unit model and the downsampled mask,

updating the original mask according to the region convolution feature information, and

in response to the updated original mask satisfying an error convergence condition, determining a target image region of the target object in the target image according to the updated original mask.

9. The apparatus according to claim 8 , wherein:

the convolution pooling layer in the second unit model comprises a first convolution pooling layer and a second convolution pooling layer;

the downsampled mask comprises a first original mask corresponding to the first convolution pooling layer and a second original mask corresponding to the second convolution pooling layer; and

when the processor is configured to cause the apparatus to perform extracting the region convolution feature information of the target image based on the convolution pooling layer in the second unit model and the downsampled mask, the processor is configured to cause the apparatus to perform:

performing convolution and pooling on the target image based on the first convolution pooling layer and the first original mask to obtain first convolution feature information, and

performing convolution and pooling on the first convolution feature information based on the second convolution pooling layer and the second original mask to obtain the region convolution feature information of the target image.

10. The apparatus according to claim 9 , wherein:

the first convolution pooling layer comprises a first convolution pooling unit and a second convolution pooling unit; and

when the processor is configured to cause the apparatus to perform performing convolution and pooling on the target image based on the first convolution pooling layer and the first original mask to obtain the first convolution feature information, the processor is configured to cause the apparatus to perform:

performing convolution and pooling on the target image based on a convolution function and a pooling function corresponding to the first convolution pooling unit to obtain input convolution feature information,

encoding the input convolution feature information based on a convolution function corresponding to the second convolution pooling unit to generate second convolution feature information,

performing pooling on feature maps of a plurality of channels of the second convolution feature information according to the first original mask separately to determine a pooling vector,

converting the pooling vector into a target vector based on an activation function corresponding to the second convolution pooling unit, and

generating the first convolution feature information according to the target vector and the first original mask, the input convolution feature information, the first convolution feature information, and the second convolution feature information having a same size.

11. The apparatus according to claim 9 , wherein when the processor is configured to cause the apparatus to perform updating the original mask according to the region convolution feature information, the processor is configured to cause the apparatus to perform:

performing deconvolution on the region convolution feature information to generate third convolution feature information;

superimposing the third convolution feature information and the first convolution feature information into target convolution feature information; and

performing deconvolution and full connection on the target convolution feature information to obtain a to-be-updated mask, and determining the to-be-updated mask as the original mask, the to-be-updated mask and the target image having a same size.

12. The apparatus according to claim 11 , wherein:

the to-be-updated mask comprises a plurality of to-be-updated mask-units; and

when the processor is configured to cause the apparatus to perform determining the to-be-updated mask as the original mask, the processor is configured to cause the apparatus to perform:

in response to a value of a to-be-updated mask-unit in the to-be-updated mask being less than or equal to a mask threshold, adjusting the value of the to-be-updated mask-unit to a first numerical value, and

in response to the value of the to-be-updated mask-unit in the to-be-updated mask being greater than the mask threshold, adjusting the value of the to-be-updated mask-unit to a second numerical value, to obtain the original mask.

13. The apparatus according to claim 8 , wherein when the processor is configured to cause the apparatus to perform determining the target image region of the target object in the target image according to the updated original mask, the processor is configured to cause the apparatus to perform:

determining the updated original mask as a target mask, the target mask comprising a plurality of target mask-units;

determining location information of a target mask-unit greater than a mask threshold in the target mask; and

using an image region corresponding to the location information in the target image as the target image region.

14. The apparatus according to claim 8 , wherein when the processor executes the instructions, the processor is configured to further cause the apparatus to perform:

obtaining a sample image comprising the target object, and obtaining a sample original mask and a sample image segmentation model, the sample image segmentation model comprising the first unit model and a sample second unit model;

downsampling the sample original mask based on the pooling layer in the first unit model to obtain a sample downsampled mask;

generating a prediction mask based on the sample second unit model and the sample downsampled mask;

determining a sample mask according to an image region of the target object in the sample image; and

training the sample second unit model according to the sample image, the prediction mask, and the sample mask to obtain the image segmentation model.

15. A non-transitory computer-readable storage medium, storing computer-readable instructions, wherein, the computer-readable instructions, when executed by a processor, are configured to cause the processor to perform:

obtaining a target image comprising a target object;

obtaining an original mask and an image segmentation model, the image segmentation model comprising a first unit model and a second unit model;

downsampling the original mask based on a pooling layer in the first unit model to obtain a downsampled mask;

extracting region convolution feature information of the target image based on a convolution pooling layer in the second unit model and the downsampled mask;

updating the original mask according to the region convolution feature information; and

in response to the updated original mask satisfying an error convergence condition, determining a target image region of the target object in the target image according to the updated original mask.

16. The non-transitory computer-readable storage medium according to claim 15 , wherein:

the convolution pooling layer in the second unit model comprises a first convolution pooling layer and a second convolution pooling layer;

the downsampled mask comprises a first original mask corresponding to the first convolution pooling layer and a second original mask corresponding to the second convolution pooling layer; and

when the computer-readable instructions are configured to cause the processor to perform extracting the region convolution feature information of the target image based on the convolution pooling layer in the second unit model and the downsampled mask, the computer-readable instructions are configured to cause the processor to perform:

performing convolution and pooling on the target image based on the first convolution pooling layer and the first original mask to obtain first convolution feature information, and

performing convolution and pooling on the first convolution feature information based on the second convolution pooling layer and the second original mask to obtain the region convolution feature information of the target image.

17. The non-transitory computer-readable storage medium according to claim 16 , wherein:

the first convolution pooling layer comprises a first convolution pooling unit and a second convolution pooling unit; and

when the computer-readable instructions are configured to cause the processor to perform performing convolution and pooling on the target image based on the first convolution pooling layer and the first original mask to obtain the first convolution feature information, the computer-readable instructions are configured to cause the processor to perform:

performing convolution and pooling on the target image based on a convolution function and a pooling function corresponding to the first convolution pooling unit to obtain input convolution feature information,

encoding the input convolution feature information based on a convolution function corresponding to the second convolution pooling unit to generate second convolution feature information,

performing pooling on feature maps of a plurality of channels of the second convolution feature information according to the first original mask separately to determine a pooling vector,

converting the pooling vector into a target vector based on an activation function corresponding to the second convolution pooling unit, and

generating the first convolution feature information according to the target vector and the first original mask, the input convolution feature information, the first convolution feature information, and the second convolution feature information having a same size.

18. The non-transitory computer-readable storage medium according to claim 16 , wherein, when the computer-readable instructions are configured to cause the processor to perform updating the original mask according to the region convolution feature information, the computer-readable instructions are configured to cause the processor to perform:

performing deconvolution on the region convolution feature information to generate third convolution feature information;

superimposing the third convolution feature information and the first convolution feature information into target convolution feature information; and

performing deconvolution and full connection on the target convolution feature information to obtain a to-be-updated mask, and determining the to-be-updated mask as the original mask, the to-be-updated mask and the target image having a same size.

19. The non-transitory computer-readable storage medium according to claim 18 , wherein:

the to-be-updated mask comprises a plurality of to-be-updated mask-units; and

when the computer-readable instructions are configured to cause the processor to perform determining the to-be-updated mask as the original mask, the computer-readable instructions are configured to cause the processor to perform:

in response to a value of a to-be-updated mask-unit in the to-be-updated mask being less than or equal to a mask threshold, adjusting the value of the to-be-updated mask-unit to a first numerical value, and

in response to the value of the to-be-updated mask-unit in the to-be-updated mask being greater than the mask threshold, adjusting the value of the to-be-updated mask-unit to a second numerical value, to obtain the original mask.

20. The non-transitory computer-readable storage medium according to claim 15 , wherein, when the computer-readable instructions are configured to cause the processor to perform determining the target image region of the target object in the target image according to the updated original mask, the computer-readable instructions are configured to cause the processor to perform:

determining the updated original mask as a target mask, the target mask comprising a plurality of target mask-units;

determining location information of a target mask-unit greater than a mask threshold in the target mask; and

using an image region corresponding to the location information in the target image as the target image region.