Image processing method and device, and storage medium
An image processing method, the method includes: obtaining a compressed image of a current image, where resolution of the current image is greater than that of the compressed image; cropping out at least one composition preview image based on the compressed image; and in response to selecting a target composition image from said at least one composition preview image displayed, cropping the current image based on a cropping range indicated by the target composition image.
1 . An image processing method, comprising:
obtaining a compressed image of a current image, wherein resolution of the current image is greater than that of the compressed image;
cropping out at least one composition preview image based on the compressed image; and
cropping, in response to selecting a target composition image from said at least one composition preview image displayed, the current image based on a cropping range indicated by the target composition image;
wherein cropping out at least one composition preview image based on the compressed image comprises:
determining a target body image type in the compressed image;
determining, based on a correspondence between a pre-constructed body image type and a composition rule, a target composition rule corresponding to the target body image type; and
obtaining at least one composition preview image by cropping the compressed image based on the target composition rule;
wherein the target composition rule comprises at least one of:
a distribution principle of a body image in the composition preview image, wherein the distribution principle comprises at least one of: a distribution principle based on an image median line, a distribution principle based on an image trisection line, a distribution principle based on a preset image quality point, a distribution principle based on a preset image inner frame region, or a principle for setting an edge or an outer frame around the body image;
a cropping circumvention principle of the body image in the composition preview image, wherein the cropping circumvention principle comprises circumventing at least one of a face, four limbs, and a specific joint of a portrait, and a face, four limbs, and a specific joint of a pet when cropping an edge of the body image; or
an inclination angle correction principle of the body image in the composition preview image.
2 . The image processing method according to claim 1 , further comprises:
executing, in response to the current image meeting an image processing condition, an operation of obtaining the compressed image of the current image;
the current image meeting the image processing condition comprises at least one of:
detecting an instruction for shooting the current image; or
detecting a trigger operation for a preset composition function entrance, wherein the preset composition function entrance is located at a large image browsing page of the current image in a photo album application.
3 . The image processing method according to claim 1 , wherein the determining a target body image type in the compressed image comprises:
recognizing the target body image type in the compressed image based on a preset image recognition algorithm.
4 . The image processing method according to claim 1 , wherein the determining a target body image type in the compressed image comprises:
determining the target body image type in the compressed image based on pre-obtained label information of the current image, wherein the pre-obtained label information comprises at least one of a face detection result of the current image and an artificial intelligence scenario detection result.
5 . The image processing method according to claim 1 , wherein the target body image type comprises a portrait type and a non-portrait type;
the portrait type further comprises a portrait subtype, wherein the portrait subtype is at least one of: single-person, double-person, or multi-person; and
the non-portrait type further comprises a non-portrait subtype, wherein the non-portrait subtype at least one of: animals, plants, scenery, constructions, or food.
6 . The image processing method according to claim 1 , wherein the target composition rule further comprises:
a preset image scale of the composition preview image.
7 . The image processing method according to claim 1 , wherein the cropping out at least one composition preview image based on the compressed image comprises:
obtaining the at least one composition preview image by inputting the compressed image to a pre-trained composition preview image generation model.
8 . The image processing method according to claim 7 , further comprising training a composition preview image generation model to form the pre-trained composition preview image generation model based on:
obtaining a plurality of sample images of different body image types;
obtaining at least one sample composition preview image of each of the plurality of sample images by cropping each of the plurality of sample images based on a composition rule corresponding to each body image type;
establishing a sample image database based on each of the plurality of sample images and the at least one sample composition preview image; and
obtaining a trained composition preview image generation model by training a pre-constructed composition preview image generation model based on the sample image database, wherein the composition preview image generation model uses a neural network model based on deep learning.
9 . The image processing method according to claim 1 , wherein
the preset image quality point is at least one of a focus of the image median line or a focus of the image trisection line, and
the preset image inner frame region is a minimum wrapped rectangle of the preset image quality point.
10 . An electronic device, comprising:
a processor, and a memory configured to store a computer program;
wherein when executing the computer program, the processor is configured to implement:
obtaining a compressed image of a current image, wherein resolution of the current image is greater than that of the compressed image;
cropping out at least one composition preview image based on the compressed image; and
cropping, in response to selecting a target composition image from said at least one composition preview image displayed, the current image based on a cropping range indicated by the target composition image;
wherein cropping out at least one composition preview image based on the compressed image comprises:
determining a target body image type in the compressed image;
determining, based on a correspondence between a pre-constructed body image type and a composition rule, a target composition rule corresponding to the target body image type; and
obtaining at least one composition preview image by cropping the compressed image based on the target composition rule;
wherein the target composition rule comprises at least one of:
a distribution principle of a body image in the composition preview image, wherein the distribution principle comprises at least one of: a distribution principle based on an image median line, a distribution principle based on an image trisection line, a distribution principle based on a preset image quality point, a distribution principle based on a preset image inner frame region, or a principle for setting an edge or an outer frame around the body image;
a cropping circumvention principle of the body image in the composition preview image, wherein the cropping circumvention principle comprises circumventing at least one of a face, four limbs, and a specific joint of a portrait, and a face, four limbs, and a specific joint of a pet when cropping an edge of the body image; or
an inclination angle correction principle of the body image in the composition preview image.
11 . The electronic device according to claim 10 , wherein when executing the computer program, the processor is further configured to implement:
executing, in response to the current image meeting an image processing condition, an operation of obtaining the compressed image of the current image;
the current image meeting the image processing condition comprises at least one of:
detecting an instruction for shooting the current image; or
detecting a trigger operation for a preset composition function entrance, wherein the preset composition function entrance is located at a large image browsing page of the current image in a photo album application.
12 . The electronic device according to claim 10 , wherein when executing the computer program, the processor is further configured to implement:
recognizing the target body image type in the compressed image based on a preset image recognition algorithm.
13 . The electronic device according to claim 10 , wherein when executing the computer program, the processor is further configured to implement:
determining the target body image type in the compressed image based on pre-obtained label information of the current image, wherein the pre-obtained label information comprises at least one of a face detection result of the current image and an artificial intelligence scenario detection result.
14 . The electronic device according to claim 10 , wherein the target body image type comprises a portrait type and a non-portrait type;
the portrait type further comprises a portrait subtype, wherein the portrait subtype is at least one of: single-person, double-person, or multi-person; and
the non-portrait type further comprises a non-portrait subtype, wherein the non-portrait subtype is at least one of:
animals, plants, scenery, constructions, or food.
15 . The electronic device according to claim 10 , wherein the target composition rule further comprises:
a preset image scale of the composition preview image.
16 . The electronic device according to claim 10 , wherein when executing the computer program, the processor is further configured to implement:
obtaining the at least one composition preview image by inputting the compressed image to a pre-trained composition preview image generation model.
17 . The electronic device according to claim 16 , wherein when executing the computer program, the processor is further configured to implement:
obtaining a plurality of sample images of different body image types;
obtaining at least one sample composition preview image of each of the plurality of sample images by cropping each of the plurality of sample images based on a composition rule corresponding to each body image type;
establishing a sample image database based on each of the plurality of sample images and the at least one sample composition preview image; and
obtaining a trained composition preview image generation model by training a pre- constructed composition preview image generation model based on the sample image database, wherein the pre-constructed composition preview image generation model uses a neural network model based on deep learning.
18 . A non-transitory computer-readable storage medium storing a computer program, the computer program when executed by a processor cause the processor to execute a method comprising:
obtaining a compressed image of a current image, wherein resolution of the current image is greater than that of the compressed image;
cropping out at least one composition preview image based on the compressed image; and
cropping, in response to selecting a target composition image from said at least one composition preview image displayed, the current image based on a cropping range indicated by the target composition image;
wherein cropping out at least one composition preview image based on the compressed image comprises:
determining a target body image type in the compressed image;
determining, based on a correspondence between a pre-constructed body image type and a composition rule, a target composition rule corresponding to the target body image type; and
obtaining at least one composition preview image by cropping the compressed image based on the target composition rule;
wherein the target composition rule comprises at least one of:
a distribution principle of a body image in the composition preview image, wherein the distribution principle comprises at least one of: a distribution principle based on an image median line, a distribution principle based on an image trisection line, a distribution principle based on a preset image quality point, a distribution principle based on a preset image inner frame region, or a principle for setting an edge or an outer frame around the body image;
a cropping circumvention principle of the body image in the composition preview image, wherein the cropping circumvention principle comprises circumventing at least one of a face, four limbs, and a specific joint of a portrait, and a face, four limbs, and a specific joint of a pet when cropping an edge of the body image; or
an inclination angle correction principle of the body image in the composition preview image.
19 . The non-transitory computer-readable storage medium according to claim 18 , wherein the method further comprises:
executing, in response to the current image meeting an image processing condition, an operation of obtaining the compressed image of the current image;
the current image meeting the image processing condition comprises at least one of:
detecting an instruction for shooting the current image; or
detecting a trigger operation for a preset composition function entrance, wherein the preset composition function entrance is located at a large image browsing page of the current image in a photo album application.