CN117115287A - Image generation method, device, electronic equipment and readable storage medium - Google Patents

Abstract

This application discloses an image generation method, device, electronic equipment and readable storage medium, which belongs to the technical field of image processing. Among them, the image generation method provided by the embodiment of the present application includes: sending a first feature map corresponding to the first image to the second electronic device; the first image includes: a foreground image and a first background image; the first feature map is used to The second electronic device generates a second feature map; receives the second feature map from the second electronic device, and generates a second background image according to the second feature map; the second background image and the first background image have different image contents; based on the foreground image and a second background image to generate a second image.

Description

Image generation method, device, electronic equipment and readable storage medium

Technical field

This application belongs to the field of image processing technology, and specifically relates to an image generation method, device, electronic equipment and readable storage medium.

Background technique

Currently, in order to avoid the leakage of user privacy, artificial intelligence generated content (AIGC) large models can be configured in electronic devices. In this way, when the user needs to generate a user image with richer content based on the captured user image, the user does not need to Instead of sending the captured user image to other devices, the electronic device can be directly triggered to input the captured user image into the AIGC large model, so that the AIGC large model can process the captured user image to generate more content that the user needs. Rich user images so that users can get the user images they need.

However, after the electronic device inputs the captured user image into the AIGC large model, the AIGC large model may require a long time of calculation to complete the processing of the captured user image to generate richer content required by the user. The user image, therefore, results in a longer time-consuming process in generating the user image required by the user, thus causing the electronic device to be less efficient in generating the image required by the user.

Contents of the invention

The purpose of the embodiments of the present application is to provide an image generation method, device, electronic device and readable storage medium, which can solve the problem of low efficiency of electronic devices in generating images required by users.

In a first aspect, embodiments of the present application provide an image generation method, applied to a first electronic device. The method includes: sending a first feature map corresponding to a first image to a second electronic device; the first image includes: a foreground image and the first background image; the first feature map is used by the second electronic device to generate a second feature map; the second feature map is received from the second electronic device, and the second background image is generated according to the second feature map; the second The background image and the first background image have different image contents; a second image is generated based on the foreground image and the second background image.

In a second aspect, embodiments of the present application provide an image generation method applied to a second electronic device. The method includes:

Receive the first feature map corresponding to the first image from the first electronic device; the first image includes: a foreground image and a first background image; input the first feature map to the first model to obtain the second feature output by the first model Figure; the second feature map is generated by processing the first feature map by the first model; sending the second feature map to the first electronic device; the second feature map is used by the first electronic device to generate a second background image, and A second image is generated based on the foreground image and the second background image; the second background image and the first background image have different image contents.

In a third aspect, embodiments of the present application provide an image generating device. The image generating device is a first image generating device. The first image generating device includes: a sending module, a receiving module and a processing module. Wherein, the sending module is used to send the first feature map corresponding to the first image to the second image generating device; the first image includes: a foreground image and a first background image; the first feature map is used for the second image generating device Generate a second feature map. A receiving module, configured to receive the second feature map from the second image generating device. A processing module configured to generate a second background image according to the second feature map received by the receiving module; the second background image and the first background image have different image contents; and generate a second image based on the foreground image and the second background image.

In a fourth aspect, embodiments of the present application provide an image generating device. The image generating device is a second image generating device. The second image generating device includes: a receiving module, a processing module and a sending module. Wherein, the receiving module is used to receive the first feature map corresponding to the first image from the first image generating device; the first image includes: a foreground image and a first background image. The processing module is configured to input the first feature map received by the receiving module into the first model to obtain the second feature map output by the first model; the second feature map is generated by processing the first feature map by the first model. A sending module, configured to send the second feature map generated by the processing module to the first image generating device; the second feature map is used by the first image generating device to generate a second background image, and based on the foreground image and the second background image, generate The second image has different image content from the first background image.

In a fifth aspect, embodiments of the present application provide an electronic device. The electronic device includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. The programs or instructions are processed by the processor. When the processor is executed, the steps of the method described in the first aspect are implemented, or the steps of the method described in the second aspect are implemented.

In a sixth aspect, embodiments of the present application provide a readable storage medium, which stores programs or instructions. When the programs or instructions are executed by a processor, the steps of the method described in the first aspect are implemented. , or implement the steps of the method described in the second aspect.

In a seventh aspect, embodiments of the present application provide a chip. The chip includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the first aspect. The steps of the method described in the second aspect, or the steps of implementing the method described in the second aspect.

In an eighth aspect, embodiments of the present application provide a computer program product. The program product is stored in a storage medium. The program product is executed by at least one processor to implement the steps of the method described in the first aspect, or to implement the method as described in the first aspect. The steps of the method described in the second aspect.

In this embodiment of the present application, the first electronic device may send a first feature map corresponding to the first image to the second electronic device. The first image includes a foreground image and a first background image. The first feature map is used for the second electronic device. The electronic device generates the second feature map so that the second electronic device can receive the first feature map and input the first feature map to the first model to obtain the first feature map generated by the first model. the second feature map, so that the second electronic device can send the second feature map to the first electronic device, so that the first electronic device can generate a second background image based on the second feature map, and the second background image and the first The background image has different image contents, and the second image is generated based on the foreground image and the second background image. Since the first electronic device can send the first feature map to the second electronic device with greater computing power, after the second electronic device inputs the first feature map into the first model, the first model only needs a shorter time The calculation can complete the processing of the first feature map to generate the second feature map, so that the first electronic device can quickly generate a second background that is different from the image content of the first background image based on the second feature map. image without the need for the first electronic device to perform calculations to generate the second background image. Therefore, the time-consuming process of generating the second background image can be reduced, so that the first electronic device can quickly generate user requirements based on the foreground image and the second background image. The second image of The second background image, therefore, can avoid the leakage of the first image or the second background image in the process of generating the second image required by the user; in this way, the first image can be improved while avoiding the leakage of user privacy. The efficiency with which an electronic device generates images required by the user.

Description of drawings

Figure 1 is one of the flow diagrams of the image generation method provided by the embodiment of the present application;

Figure 2 is the second schematic flowchart of the image generation method provided by the embodiment of the present application;

Figure 3 is the third schematic flowchart of the image generation method provided by the embodiment of the present application;

Figure 4 is the fourth schematic flowchart of the image generation method provided by the embodiment of the present application;

Figure 5 is a schematic structural diagram of a first image generation device provided by an embodiment of the present application;

Figure 6 is a schematic structural diagram of a second image generation device provided by an embodiment of the present application;

Figure 7 is one of the schematic diagrams of the hardware structure of the electronic device provided by the embodiment of the present application;

Figure 8 is a second schematic diagram of the hardware structure of the electronic device provided by the embodiment of the present application;

Figure 9 is a schematic diagram of the hardware structure of a server provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art fall within the scope of protection of this application.

The terms "first", "second", etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the figures so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in orders other than those illustrated or described herein, and that "first," "second," etc. are distinguished Objects are usually of one type, and the number of objects is not limited. For example, the first image may be one or multiple. In addition, "and/or" in the description and claims indicates at least one of the connected objects, and the character "/" generally indicates that the related objects are in an "or" relationship.

The terms "at least one (item)", "at least one", etc. in the description and claims of this application refer to any one, any two or a combination of more than two of the included objects. For example, at least one (item) of a, b, c can represent: "a", "b", "c", "a and b", "a and c", "b and c" and "a , b and c", where a, b, c can be single or multiple. Similarly, "at least two (items)" refers to two or more than two, and its meaning is similar to "at least one (item)".

The image generation method, device, electronic device and readable storage medium provided by the embodiments of the present application will be described in detail through specific embodiments and application scenarios with reference to the accompanying drawings.

For the image generation method provided by the embodiments of the present application, the execution subject may be the first image generation device and the second image generation device, or the first electronic device and the second electronic device, or the functions in the first electronic device and the second electronic device. module or entity. In the embodiment of the present application, the image generation method performed by the first electronic device and the second electronic device is used as an example to illustrate the image generation method provided by the embodiment of the present application.

Figure 1 shows a flow chart of an image generation method provided by an embodiment of the present application. As shown in Figure 1, the image generation method provided by the embodiment of the present application may include the following steps 101 to 106.

Step 101: The first electronic device sends the first feature map corresponding to the first image to the second electronic device.

In some embodiments of the present application, the first electronic device may be an electronic device used by a user. For example, the first electronic device may be a mobile phone, a computer, a tablet, etc. The above-mentioned second electronic device may be any of the following: electronic devices used by other users, servers, etc.

In some embodiments of the present application, the computing power of the second electronic device is greater than the computing power of the first electronic device.

In some embodiments of the present application, the above-mentioned first image may be an image captured by the first electronic device through a camera, or an image received from other devices. For example, the first image may be a 4K image captured by the first electronic device through a camera.

In the embodiment of the present application, the above-mentioned first image includes: a foreground image and a first background image; the above-mentioned first feature map is used by the second electronic device to generate a second feature map.

It can be understood that the first image is composed of a foreground image and a first background image.

In some embodiments of the present application, the foreground image may include at least one of the following: a person image and a foreground object image. It should be noted that the above-mentioned "person image" can be understood as: the image area where the person in the foreground image is located. The above-mentioned "foreground object image" can be understood as: the image area where the object in the foreground image is located, and the object may include at least one of the following: animals, scenery, etc.

In some embodiments of the present application, the above-mentioned first feature map may be a feature map obtained by extracting features from the first image by the first electronic device through the second model. Wherein, the second model may be a convolutional network model.

The following will illustrate a specific solution for the first electronic device to obtain the first feature map.

In some embodiments of the present application, as shown in Figure 2 in conjunction with Figure 1, before the above step 101, the image generation method provided by the embodiment of the present application may also include the following step 201.

Step 201: The first electronic device inputs the first image into the encoder to obtain the first feature map output by the encoder.

In the embodiment of the present application, the above-mentioned first feature map is obtained by encoding the first image by the encoder.

In some embodiments of the present application, the above-mentioned encoder may specifically be: a variational autoencoder (VAE) in a convolutional network model.

Optionally, the above VAE may be obtained by training a preset convolutional network model by the first electronic device. It can be understood that the VAE in the first electronic device is different from the VAE in other devices, that is, other devices cannot decode and obtain visible images through VAE.

In some embodiments of the present application, the size of the above-mentioned first feature map is smaller than the first image. It can be understood that the amount of data required to transmit the first feature map is smaller than the amount of data required to transmit the first image.

In the embodiment of the present application, the first electronic device can encode the first image into latent space features, that is, the first feature map, through VAE. The first feature map cannot be decoded into a visible image on other devices. Therefore, , which can avoid user privacy leakage.

It can be seen that since the first electronic device can encode the first image into a first feature map of a certain size through the encoder, after sending the first feature map to the second electronic device, the first feature map cannot be displayed on the second electronic device. The first electronic device can be decoded into a visible image on any one of the devices and other devices. Therefore, the user's privacy can be avoided while saving transmission resources. In this way, the first electronic device can be improved while improving transmission efficiency. security.

In some embodiments of the present application, the above-mentioned second feature map may specifically be a feature map corresponding to other background images. Wherein, the other background image may be the second background image in the following embodiments. The image content of the other background image is different from that of the first background image. The image content may include at least one of the following: image color bar, image style, Image scenery, etc., the image style may include at least one of the following: image color, image tone, content layout, etc.

It can be understood that the other background images have similar image structures to the first background image, and the content of the other background images is richer.

In some embodiments of the present application, when the user needs to generate a second multimedia file with richer content based on the first multimedia file, the user can trigger the first electronic device to select the first multimedia file to The first electronic device can obtain the first feature map corresponding to the first image, and send the first feature map to the second electronic device through the first application program.

Optionally, the first multimedia file may be any of the following: a first image or a first video. In the case where the first multimedia file is a first video, the first image may be a first Any one of the multiple frames of the video. The first multimedia file may be a multimedia file captured by the camera of the first electronic device, or a multimedia file received from other devices.

Optionally, the second multimedia file may be any of the following: other images or a second video. The other images may be the second images in the following embodiments, where the second multimedia file is the second image. In the case of a second video, the other image may be an image among multiple frames of the second video that has the same playback timestamp corresponding to the first image.

Optionally, the first application may be any of the following: a chatting application, an image processing application, a shooting application, etc.

In some embodiments of the present application, the first electronic device may also send first information to the second electronic device, where the first information is used to generate a second feature map together with the first feature map. The first information may include at least one of the following: a depth map corresponding to the first image and first text data; for example, the first information includes a depth map corresponding to the first image and first text data. It can be understood that the second electronic device can generate the second feature map based on the first information and the first feature map.

Optionally, the first information is used by the second electronic device to determine the image content of the other background images.

Optionally, in the case where the first image is an image captured by the first electronic device through a camera, when capturing the first image, the first electronic device may also capture a time of flight (TOF) of the first electronic device. The above depth map collected by the sensor Sensor. In the case where the first image is an image received by the first electronic device from another device, the first electronic device may also receive the above-mentioned depth map from the other device.

Optionally, the above-mentioned first text data may specifically be text data input by the user in the first electronic device.

Step 102: The second electronic device receives the first feature map corresponding to the first image from the first electronic device.

In some embodiments of the present application, the computing power of the second electronic device is greater than the computing power of the first electronic device.

In this embodiment of the present application, the above-mentioned first image includes: a foreground image and a first background image.

Step 103: The second electronic device inputs the first feature map to the first model to obtain the second feature map output by the first model.

In the embodiment of the present application, the above-mentioned second feature map is generated by processing the first feature map by the first model.

It can be understood that since the computing power of the second electronic device is greater than that of the first electronic device, after the second electronic device inputs the first feature map to the first model, the first model can quickly calculate the first feature map. Processing is performed to quickly generate and output the second feature map.

In some embodiments of the present application, the above-mentioned first model may specifically be: an artificial intelligence generated content (Artificial Intelligence Generated Content, AIGC) large model.

It should be noted that, for the description of processing the first feature map to generate the second feature map for the AIGC large model, reference can be made to the specific description in the related art, and the embodiments of the present application will not be repeated here.

In one example, the second electronic device can directly input the first feature map to the first model to obtain the second feature map output by the first model.

In another example, the second electronic device may first obtain the feature area corresponding to the foreground image in the first feature map in the first image, and obtain the feature area corresponding to the foreground image in the first feature map. A feature area corresponding to a background image in the first feature map is then input into the first model to obtain a second feature map output by the first model.

Optionally, the second electronic device may use a Salient Object Detection (SOD) algorithm to detect the first feature map to determine the corresponding feature area of the foreground image in the first feature map. Alternatively, the second electronic device may receive a piece of information from the first electronic device indicating the position of the person image in the first image, so that the second electronic device can determine the corresponding position of the foreground image in the first feature map based on the information. Characteristic area.

It should be noted that for the description of the above-mentioned SOD algorithm, reference may be made to the specific descriptions in related technologies, and the embodiments of the present application will not be repeated here.

In some embodiments of the present application, the second electronic device can also receive the first information from the first electronic device, so that the second electronic device can input the first feature map and the first information to the first model to obtain the first model. The second feature map of the output. Wherein, the first information may include at least one of the following: a depth map corresponding to the first image, and first text data. The first information is used by the first model to determine the image content of the other background images; for example, the third One piece of information includes a depth map corresponding to the first image and first text data.

Optionally, when the first information does not include the depth map corresponding to the first image, the second electronic device may first use a monocular depth estimation module to detect the first feature map to obtain the depth corresponding to the first image. image, and then input the first feature map, the depth map corresponding to the first image, and the first text data into the first model to obtain the second feature map.

It should be noted that for the description of the depth map and the first text data, reference may be made to the specific descriptions in the above embodiments, and the embodiments of the present application will not be repeated here.

Step 104: The second electronic device sends the second feature map to the first electronic device.

In the embodiment of the present application, the above-mentioned second feature map is used by the first electronic device to generate a second background image, and generate a second image based on the foreground image and the second background image. The second background image and the first background image are The content is different.

Step 105: The first electronic device receives the second feature map from the second electronic device, and generates a second background image based on the second feature map.

In this embodiment of the present application, the image content of the second background image and the first background image are different.

In some embodiments of the present application, the second background image may be an image generated by the first electronic device by processing the second feature map through the second model.

A specific solution for the first electronic device to generate the second background image will be described below with an example.

In some embodiments of the present application, as shown in FIG. 3 in conjunction with FIG. 1 , the above step 105 may be implemented through the following step 105a.

Step 105a: The first electronic device receives the second feature map from the second electronic device, and inputs the second feature map into the decoder of the first electronic device to obtain a second background image output by the decoder.

In the embodiment of the present application, the above-mentioned second background image is obtained by decoding the second feature map by the decoder.

In some embodiments of the present application, the above-mentioned decoder may be the above-mentioned VAE.

In some embodiments of the present application, the size of the above-mentioned second feature map is smaller than the second background image. It can be understood that the amount of data required to transmit the second feature map is smaller than the amount of data required to transmit the second background image.

It can be seen that since only the first electronic device can decode the second feature map into a second background image through the decoder, but the second electronic device and other devices cannot decode the second feature map into a visible image, therefore, it can be avoided The user's privacy is leaked, thus, the security of the first electronic device can be improved.

Step 106: The first electronic device generates a second image based on the foreground image and the second background image.

In some embodiments of the present application, the first electronic device may fuse the foreground image and the second background image to generate the second image.

Optionally, the first electronic device may superimpose the foreground image on the second background image to fuse the foreground image and the second background image.

In the embodiment of the present application, since the characters and objects in the foreground image of the first image are often the characters and objects required by the user, that is, the user may not require the characters and objects in the foreground image to change. Therefore, the first electronic device can Based on the foreground image and the second background image, a second image is generated, and the characters and objects in the second image are exactly the same as the characters and objects in the foreground image in the first image.

Embodiments of the present application provide an image generation method. A first electronic device can send a first feature map corresponding to a first image to a second electronic device. The first image includes a foreground image and a first background image. The first feature map Used by the second electronic device to generate the second feature map, so that the second electronic device can receive the first feature map and input the first feature map to the first model to obtain the first model's response to the first feature map The generated second feature map is processed, so that the second electronic device can send the second feature map to the first electronic device, so that the first electronic device can generate a second background image according to the second feature map, and the second background image The image content of the image and the first background image are different, and the second image is generated based on the foreground image and the second background image. Since the first electronic device can send the first feature map to the second electronic device with greater computing power, after the second electronic device inputs the first feature map into the first model, the first model only needs a shorter time The calculation can complete the processing of the first feature map to generate the second feature map, so that the first electronic device can quickly generate a second background that is different from the image content of the first background image based on the second feature map. image without the need for the first electronic device to perform calculations to generate the second background image. Therefore, the time-consuming process of generating the second background image can be reduced, so that the first electronic device can quickly generate user requirements based on the foreground image and the second background image. The second image of Therefore, it is possible to avoid the leakage of the first image or the second background image in the process of generating the second image required by the user; in this way, the first electronic image can be improved while avoiding the leakage of the user's privacy. The efficiency with which the device generates the images required by the user.

The following will illustrate a specific solution for the second electronic device to generate the second feature map.

In some embodiments of the present application, the foreground image includes a character image. Optionally, in conjunction with Figure 1, as shown in Figure 4, before the above step 103, the image generation method provided by the embodiment of the present application may also include the following steps 301 to 304, and the above step 103 may be specifically performed by the following Steps 103a and 103b are implemented.

Step 301: The first electronic device determines the first mask image based on the first image.

It should be noted that the embodiment of the present application does not limit the execution order of step 101 and step 301 here. In one example, step 101 can be performed first, and then step 301 can be performed; in another example, step 301 can be performed first, and then step 101 can be performed; in yet another example, while step 101 can be performed, Execute step 301. In Figure 4, step 101 is executed first, and then step 301 is executed.

In the embodiment of the present application, the above-mentioned first mask image indicates the position of the character image in the first image.

In some embodiments of the present application, each pixel point in the first mask image corresponds to a pixel point in the first image, and in the first mask image, each pixel point in the first mask image corresponds to a pixel point in the person image. The pixel value of the pixel point can be "1", and the pixel value of the pixel point corresponding to the pixel point in other images can be "0".

In some embodiments of the present application, the first electronic device may use the portrait segmentation algorithm module to calculate the first image to determine the first mask image.

It should be noted that for the description of the above portrait segmentation algorithm, reference may be made to the specific descriptions in related technologies, and the embodiments of the present application will not be repeated here.

Step 302: The first electronic device sends the first mask image to the second electronic device.

In this embodiment of the present application, the above-mentioned first mask image is used by the second electronic device to determine the first feature area corresponding to the foreground image in the first feature map.

In the embodiment of the present application, the above-mentioned second feature map is generated by the second electronic device based on the second feature area corresponding to the first background image in the first feature map. The second feature area is generated by the second electronic device based on the first feature. area determined.

In the embodiment of the present application, due to the limited computing power of the first electronic device, it may take a long time for the first electronic device to determine the position of the foreground image in the first image. Therefore, the first electronic device can quickly determine the position of the foreground image in the first image. The first mask image is sent to the second electronic device, so that the second electronic device with greater computing power can determine the first feature area corresponding to the foreground image in the first feature map.

Step 303: The second electronic device receives the first mask image from the first electronic device.

In the embodiment of the present application, the above-mentioned first mask image indicates the position of the character image in the first image.

Step 304: The second electronic device determines the first feature area corresponding to the foreground image in the first feature map based on the first mask image.

In some embodiments of the present application, the foreground image may include a person image and a foreground object image, so that the second electronic device may first determine the characteristic area corresponding to the person image in the first feature map based on the first mask image, and then The first feature map is then detected to determine the corresponding feature area of the foreground object image in the first feature map, so that the second electronic device can determine the corresponding feature area of the person image and the foreground object image in the first feature map. The first feature area corresponding to the foreground image in the first feature map.

In some embodiments of the present application, the foreground image also includes a foreground object image. Optionally, the above step 304 can be specifically implemented through the following steps 304a and 304b.

Step 304a: The second electronic device determines the third feature area corresponding to the person image in the first feature map from the first feature map based on the first mask image.

In some embodiments of the present application, the second electronic device may first determine the position of the corresponding pixel in the first feature map based on the position of the pixel with a pixel value of “1” in the first mask image, and then The area where the corresponding pixel point in the first feature map is located is determined as the third feature area.

Step 304b: The second electronic device determines the third feature area and the fourth feature area corresponding to the foreground object image in the first feature map as the first feature area.

In some embodiments of the present application, the second electronic device may first use the foreground segmentation module to detect the first feature map using the SOD algorithm to determine the fourth feature area, and then combine the third feature area and the fourth feature The area constituted by the area is determined as the first characteristic area.

It can be seen from this that since the second electronic device can accurately determine the third feature area corresponding to the person image in the foreground image in the first feature map from the first feature map based on the first mask image, the electronic device can thus determine based on the first feature map. The third feature area and the fourth feature area corresponding to the foreground image in the first feature map accurately determine the first feature area corresponding to the foreground image in the first feature map.

Step 103a: The second electronic device obtains the second feature area except the first feature area from the first feature map.

In the embodiment of the present application, the above-mentioned second feature area is the corresponding feature area of the first background image in the first feature map.

In some embodiments of the present application, the second electronic device may first set the feature value of each feature point in the first feature area to "0", and then randomly set the feature value of each feature point in the first feature area. value, so that the second characteristic area can be obtained.

Step 103b: The second electronic device inputs the second feature region into the first model to obtain a second feature map output by the first model.

In the embodiment of the present application, in order to avoid the presence of people and objects in the foreground image of the first image in the second background image, and thereby avoid ghost images in the second image, the first electronic device can use the first mask image to The second electronic device indicates the position of the person image in the foreground image in the first image, so that the second electronic device can determine the first feature area corresponding to the foreground image in the first feature map based on the position, and obtain the corresponding first feature area from the first feature map. The first feature area is deleted from the feature map, so that the feature area corresponding to the foreground image does not exist in the second feature map generated by the second electronic device through the first model, and further, the person and person in the foreground image do not exist in the second background image. object.

It can be seen from this that the first electronic device can send the first mask image to the second electronic device, so that the second electronic device can determine the first feature area corresponding to the foreground image in the first feature map, and obtain the first feature map from the first feature map. Obtain the second feature area except the first feature area. Therefore, after the second electronic device inputs the second feature area to the first model, a second feature map that does not include the feature area corresponding to the foreground image can be obtained, that is, the second feature map can be obtained. The characters and objects in the foreground image do not exist in the second background image generated by an electronic device. Therefore, ghost images in the generated second image can be avoided; and further, since the first electronic device and the second The electronic device determines the first feature area corresponding to the foreground image in the first feature map instead of only determining it through the first electronic device. Therefore, the time-consuming process of determining the first feature area can be reduced, thereby reducing the generation of the second feature area. The time consuming of the feature map; in this way, the quality of the second image generated by the first electronic device required by the user can be further improved, while the time consuming of generating the second image can be reduced.

In the embodiment of the present application, after the second electronic device receives the first mask image, the second electronic device can also update the first mask image according to the feature area corresponding to the foreground object image in the first feature map. The updated The first mask image indicates the position of the foreground image in the first image, so that the first electronic device can accurately fuse the foreground image and the second background image based on the updated first mask image to obtain the third image required by the user. Two images, examples will be given below.

In some embodiments of the present application, the foreground image also includes a foreground object image. Optionally, after the above step 303, the image generation method provided by the embodiment of the present application may also include the following steps 401 to 404, and the above step 106 may be specifically implemented through the following step 106a.

Step 401: The second electronic device determines a third mask image based on the fourth feature area corresponding to the foreground object image in the first feature map.

In the embodiment of the present application, the above-mentioned third mask image indicates the position of the foreground object image in the first image.

It should be noted that, for the description of determining the fourth characteristic area by the second electronic device, reference may be made to the specific description in the above embodiments, and the embodiments of the present application will not be repeated here.

In some embodiments of the present application, each pixel point in the third mask image corresponds to a pixel point in the first image, and in the third mask image, each pixel point in the foreground object image corresponds to The pixel value of the corresponding pixel point may be "1", and the pixel value of the pixel point corresponding to the pixel point in other images may be "0".

In some embodiments of the present application, the second electronic device can determine the location of the corresponding pixel point in the first image based on the location of each feature point in the fourth feature area, and then determine the location of the corresponding pixel point based on the area where the corresponding pixel point is located. position, determine the position of the foreground object image in the first image, and generate a third mask image based on the position.

Step 402: The second electronic device generates a second mask image based on the first mask image and the third mask image.

In this embodiment of the present application, the above-mentioned second mask image indicates the position of the foreground image in the first image.

In some embodiments of the present application, the second electronic device may superimpose the first mask image and the second mask image to generate the second mask image.

Step 403: The second electronic device sends the second mask image to the first electronic device.

In the embodiment of the present application, the above-mentioned second mask image is used by the first electronic device to fuse the foreground image and the second background image to obtain the second image.

Step 404: The first electronic device receives the second mask image from the second electronic device.

In the embodiment of the present application, the above-mentioned second mask image indicates the position of the foreground image in the first image, and the second mask image is determined by the second electronic device based on the first mask image.

Step 106a: The first electronic device fuses the foreground image and the second background image based on the second mask image to obtain a second image.

In some embodiments of the present application, the first electronic device may first multiply the second mask image and the first image to obtain a third image. The third image only includes a foreground image, and the third image has The size is the same as that of the first image, so that the first electronic device can directly superimpose the third image on the second background image to generate the second image.

It can be seen from this that the second electronic device can also determine the third mask image indicating the position of the foreground image in the first image, and generate a third mask image indicating the position of the foreground image in the first image based on the third mask image and the first mask image. A position in an image, so that after the first electronic device fuses the foreground image and the second background image based on the second mask image, the position of the foreground image in the second image can be the same as the position in the first image. Therefore, it can be avoided that the position of the foreground image in the two images does not match, causing the composition of the second image to change, so that the user does not need to adjust the second image again. In this way, the generation of the first electronic device can be further improved. Efficiency of images required by users.

For the image generation method provided by the embodiments of the present application, the execution subject may be an image generation device. In the embodiments of the present application, an image generation device executing an image generation method is used as an example to illustrate the image generation device provided by the embodiments of the present application.

FIG. 5 shows a possible structural schematic diagram of the image generation device involved in the embodiment of the present application. The image generation device is a first image generation device. As shown in FIG. 5 , the first image generation device 50 provided by the embodiment of the present application may include: a sending module 51 , a receiving module 52 and a processing module 53 .

Among them, the sending module 51 is used to send the first feature map corresponding to the first image to the second image generation device; the first image includes: a foreground image and a first background image; the first feature map is used to generate the second image The device generates a second feature map. The receiving module 52 is configured to receive the second feature map from the second image generating device. The processing module 53 is configured to generate a second background image according to the second feature map received by the receiving module 52; the image contents of the second background image and the first background image are different; and based on the foreground image and the second background image, generate a second background image. image.

Embodiments of the present application provide an image generation device. Since the first image generation device can send a first feature map to a second image generation device with greater computing power, the second image generation device inputs the first feature map to After the first model, the first model only needs a short time of calculation to complete the processing of the first feature map to generate the second feature map, so that the first image generating device can quickly generate the second feature map based on the second feature map. Generate a second background image that is different in image content from the first background image without requiring the first image generating device to perform calculations to generate the second background image. Therefore, the time-consuming process of generating the second background image can be reduced, so that the first image The generating device can quickly generate the second image required by the user based on the foreground image and the second background image; and, because in the process of generating the second image required by the user, between the first image generating device and the second image generating device The transmission between is the first feature map and the second feature map, rather than the first image and the second background image. Therefore, the leakage of the first image or the second background image during the process of generating the second image required by the user can be avoided. situation; in this way, the efficiency of the first image generating device in generating images required by the user can be improved while avoiding leakage of user privacy.

In a possible implementation, the foreground image includes a person image. The above-mentioned processing module 53 is also used to determine the first mask image based on the first image before the receiving module 52 receives the second feature map from the second image generation device; the first mask image indicates that the character image is in the first image. location in. The above-mentioned sending module 51 is also used to send the first mask image determined by the processing module 53 to the second image generation device; the first mask image is used by the second image generation device to determine the corresponding position of the foreground image in the first feature map. The first characteristic area. Wherein, the above-mentioned second feature map is generated by the second image generating device based on the second feature area corresponding to the first background image in the first feature map, and the second feature area is determined by the second image generating device based on the first feature area. of.

In a possible implementation, the above-mentioned receiving module 52 is also configured to receive the second mask image from the second image generating device after the sending module 51 sends the first mask image to the second image generating device; The second mask image indicates the position of the foreground image in the first image, and the second mask image is determined by the second image generating device based on the first mask image. The above-mentioned processing module 53 is specifically used to fuse the foreground image and the second background image based on the second mask image received by the receiving module to obtain the second image.

In a possible implementation, the above-mentioned processing module 53 is also used to input the first image to the first image generation device before the sending module 51 sends the first feature map corresponding to the first image to the second image generation device. In the encoder of 50, the first feature map output by the encoder is obtained; the first feature map is obtained by encoding the first image by the encoder.

In a possible implementation, the above-mentioned processing module 53 is specifically configured to input the second feature map into the decoder of the first image generation device 50 to obtain a second background image output by the decoder; the second background image It is obtained by decoding the second feature map by the decoder.

The image generating device in the embodiment of the present application may be an electronic device or a component in the electronic device, such as an integrated circuit or a chip. The electronic device may be a terminal or other devices other than the terminal. Illustratively, the electronic device may be a mobile phone, a tablet computer, a notebook computer, a handheld computer, a vehicle-mounted electronic device, a mobile internet device (MID), or an augmented reality (AR)/virtual reality (VR) device. , robots, wearable devices, ultra-mobile personal computers (UMPC), netbooks or personal digital assistants (PDA), etc., and can also be used for servers, network attached storage (NAS), personal Computer (personal computer, PC), television (television, TV), teller machine or self-service machine, etc. are not specifically limited in the embodiments of this application.

The image generation device in the embodiment of the present application may be a device with an operating system. The operating system can be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiments of this application.

The image generation device provided by the embodiments of the present application can implement various processes implemented by the method embodiments of Figures 1 to 4. To avoid repetition, they will not be described again here.

FIG. 6 shows a possible structural diagram of the image generation device involved in the embodiment of the present application, where the image generation device is a second image generation device. As shown in FIG. 6 , the second image generation device 60 provided by the embodiment of the present application may include: a receiving module 61 , a processing module 62 and a sending module 63 .

Among them, the receiving module 61 is used to receive the first feature map corresponding to the first image from the first image generating device; the first image includes: a foreground image and a first background image. The processing module 62 is used to input the first feature map received by the receiving module 61 into the first model to obtain the second feature map output by the first model; the second feature map is generated by processing the first feature map by the first model. of. The sending module 63 is used to send the second feature map generated by the processing module 62 to the first image generating device; the second feature map is used by the first image generating device to generate a second background image, and is based on the foreground image and the second background image. , generate a second image, the second background image and the first background image have different image contents.

Embodiments of the present application provide an image generation device. Since the first image generation device can send a first feature map to a second image generation device with greater computing power, the second image generation device inputs the first feature map to After the first model, the first model only needs a short time of calculation to complete the processing of the first feature map to generate the second feature map, so that the first image generating device can quickly generate the second feature map based on the second feature map. Generate a second background image that is different in image content from the first background image without requiring the first image generating device to perform calculations to generate the second background image. Therefore, the time-consuming process of generating the second background image can be reduced, so that the first image The generating device can quickly generate the second image required by the user based on the foreground image and the second background image; and, because in the process of generating the second image required by the user, between the first image generating device and the second image generating device The transmission between is the first feature map and the second feature map, rather than the first image and the second background image. Therefore, the leakage of the first image or the second background image during the process of generating the second image required by the user can be avoided. situation; in this way, the efficiency of the first image generating device in generating images required by the user can be improved while avoiding leakage of user privacy.

In a possible implementation, the foreground image includes a person image. The above-mentioned receiving module 61 is also used to receive the first mask image from the first image generation device before the processing module 62 inputs the first feature map to the first model and obtains the second feature map output by the first model; A mask image indicates the position of the person image in the first image. The above-mentioned processing module 62 is also used to determine the first feature area corresponding to the foreground image in the first feature map based on the first mask image received by the receiving module 61 . The above-mentioned processing module 62 is specifically used to obtain a second feature area other than the first feature area from the first feature map. The second feature area is the corresponding feature area of the first background image in the first feature map, and The second feature region is input to the first model to obtain a second feature map output by the first model.

In a possible implementation, the foreground image also includes a foreground object image. The above-mentioned processing module 62 is specifically used to determine the third feature area corresponding to the person image in the first feature map from the first feature map according to the first mask image; and combine the third feature area and the foreground object image in the first feature map. The corresponding fourth feature area in the feature map is determined as the first feature area.

In a possible implementation, the foreground image also includes a foreground object image. The above-mentioned processing module 62 is also configured to determine the third mask image according to the fourth feature area corresponding to the foreground object image in the first feature map after the receiving module 61 receives the first mask image from the first image generation device; The third mask image indicates the position of the foreground image in the first image; and a second mask image is generated based on the first mask image and the third mask image; the second mask image indicates the position of the foreground image in the first image. A location in an image. The above-mentioned sending module 63 is also used to send the second mask image generated by the processing module 62 to the first image generating device; the second mask image is used by the first image generating device to fuse the foreground image and the second background image, Get the second image.

The image generating device in the embodiment of the present application may be an electronic device or a component in the electronic device, such as an integrated circuit or a chip. The electronic device may be a terminal or other devices other than the terminal. Illustratively, the electronic device may be a mobile phone, a tablet computer, a notebook computer, a handheld computer, a vehicle-mounted electronic device, a mobile internet device (MID), or an augmented reality (AR)/virtual reality (VR) device. , robots, wearable devices, ultra-mobile personal computers (UMPC), netbooks or personal digital assistants (PDA), etc., and can also be used for servers, network attached storage (NAS), personal Computer (personal computer, PC), television (television, TV), teller machine or self-service machine, etc. are not specifically limited in the embodiments of this application.

The image generation device in the embodiment of the present application may be a device with an operating system. The operating system can be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiments of this application.

The image generation device provided by the embodiments of the present application can implement various processes implemented by the method embodiments of Figures 1 to 4. To avoid repetition, they will not be described again here.

In some implementations of the present application, as shown in Figure 7, the embodiment of the present application also provides an electronic device 70, including a processor 71 and a memory 72. The memory 72 stores programs that can run on the processor 71. or instructions. When executed by the processor 71, the program or instructions implement each process step of the above image generation method embodiment and can achieve the same technical effect. To avoid duplication, they will not be described again here.

It should be noted that the electronic devices in the embodiments of the present application include the above-mentioned mobile electronic devices and non-mobile electronic devices.

FIG. 8 is a schematic diagram of the hardware structure of an electronic device that implements an embodiment of the present application. The electronic device is a first electronic device.

The electronic device 100 includes but is not limited to: radio frequency unit 101, network module 102, audio output unit 103, input unit 104, sensor 105, display unit 106, user input unit 107, interface unit 108, memory 109, processor 110, etc. part.

Those skilled in the art can understand that the electronic device 100 may also include a power supply (such as a battery) that supplies power to various components. The power supply may be logically connected to the processor 110 through a power management system, thereby managing charging, discharging, and function through the power management system. Consumption management and other functions. The structure of the electronic device shown in Figure 8 does not constitute a limitation on the electronic device. The electronic device may include more or less components than shown in the figure, or combine certain components, or arrange different components, which will not be described again here. .

Among them, the radio frequency unit 101 is used to send the first feature map corresponding to the first image to the second electronic device; the first image includes: a foreground image and a first background image; the first feature map is used for the second electronic device to generate a second feature map; and receiving the second feature map from the second electronic device, and generating a second background image based on the second feature map; the image content of the second background image and the first background image are different.

The processor 110 is configured to generate a second image based on the foreground image and the second background image.

Embodiments of the present application provide an electronic device. Since the first electronic device can send a first feature map to a second electronic device with greater computing power, after the second electronic device inputs the first feature map into the first model , the first model only needs a short calculation time to complete the processing of the first feature map to generate the second feature map, so that the first electronic device can quickly generate the first feature map based on the second feature map. A second background image with different image content does not require the first electronic device to perform calculations to generate the second background image. Therefore, the time-consuming process of generating the second background image can be reduced, so that the first electronic device can quickly generate the second background image based on the foreground image. The image and the second background image generate a second image required by the user; and, during the process of generating the second image required by the user, the first feature map and the second electronic device are transmitted between the first electronic device and the second electronic device. two feature maps instead of the first image and the second background image. Therefore, the leakage of the first image or the second background image in the process of generating the second image required by the user can be avoided; in this way, the user can avoid the leakage of the first image or the second background image. In the case of privacy leakage, the efficiency of the first electronic device in generating images required by the user is improved.

In some embodiments of the present application, the foreground image includes a character image.

The processor 110 is further configured to determine a first mask image according to the first image; the first mask image indicates the position of the person image in the first image.

The radio frequency unit 101 is also configured to send a first mask image to the second electronic device; the first mask image is used by the second electronic device to determine the first feature area corresponding to the foreground image in the first feature map.

Wherein, the above-mentioned second feature map is generated by the second electronic device based on the second feature area corresponding to the first background image in the first feature map, and the second feature area is determined by the second electronic device based on the first feature area.

In some embodiments of the present application, the radio frequency unit 101 is also used to receive a second mask image from a second electronic device; the second mask image indicates the position of the foreground image in the first image, and the second mask image The image is determined by the second electronic device based on the first mask image.

The processor 110 is specifically configured to fuse the foreground image and the second background image based on the second mask image to obtain the second image.

In some embodiments of the present application, the processor 110 is also configured to input the first image into the encoder of the first electronic device to obtain the first feature map output by the encoder; the first feature map is the encoder pair The first image is encoded.

In some embodiments of the present application, the processor 110 is specifically configured to input the second feature map into the decoder of the first electronic device to obtain a second background image output by the decoder; the second background image is the decoder Obtained by decoding the second feature map.

It should be understood that in the embodiment of the present application, the input unit 104 may include a graphics processing unit (GPU) 1041 and a microphone 1042. The graphics processor 1041 is responsible for the image capture device (such as Process the image data of still pictures or videos obtained by the camera). The display unit 106 may include a display panel 1061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like. The user input unit 107 includes a touch panel 1071 and at least one of other input devices 1072 . Touch panel 1071 is also called a touch screen. The touch panel 1071 may include two parts: a touch detection device and a touch controller. Other input devices 1072 may include, but are not limited to, physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.

Memory 109 may be used to store software programs as well as various data. The memory 109 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, Image playback function, etc.) etc. Additionally, memory 109 may include volatile memory or nonvolatile memory, or memory 109 may include both volatile and nonvolatile memory. Among them, the non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically removable memory. Erase programmable read-only memory (electrically EPROM, EEPROM) or flash memory. Volatile memory can be random access memory (RAM), static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (doubledatarate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (synch link DRAM, SLDRAM) ) and direct memory bus random access memory (directrambus RAM, DRRAM). Memory 109 in embodiments of the present application includes, but is not limited to, these and any other suitable types of memory.

The processor 110 may include one or more processing units; optionally, the processor 110 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 110 .

Figure 9 is a schematic diagram of the hardware structure of a server that implements an embodiment of the present application.

The server can implement the details of the image generation method corresponding to the second electronic device in the embodiment shown in Figures 1 to 4, and achieve the same effect. As shown in FIG. 9 , the server 80 includes: a processor 81 , a memory 82 and a computer program stored on the memory 82 and executable on the processor. The various components in the server 80 are coupled together through a bus system 83 . It can be understood that the bus system 83 is used to implement connection communication between these components. When the computer program is executed by the processor, it implements each process step of the image generation method in the above embodiment, and can achieve the same technical effect. To avoid duplication, it will not be described again here.

Embodiments of the present application also provide a readable storage medium. Programs or instructions are stored on the readable storage medium. When the program or instructions are executed by a processor, each process of the above image generation method embodiment is implemented, and the same can be achieved. The technical effects will not be repeated here to avoid repetition.

Wherein, the processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.

An embodiment of the present application further provides a chip. The chip includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the above image generation method embodiment. Each process can achieve the same technical effect. To avoid duplication, it will not be described again here.

It should be understood that the chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-a-chip or system-on-chip, etc.

Embodiments of the present application provide a computer program product. The program product is stored in a storage medium. The program product is executed by at least one processor to implement each process of the above image generation method embodiment, and can achieve the same technical effect. , to avoid repetition, will not be repeated here.

It should be noted that, in this document, the terms "comprising", "comprises" or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or device that includes a series of elements not only includes those elements, It also includes other elements not expressly listed or inherent in the process, method, article or apparatus. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article or apparatus that includes that element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, but may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. Functions may be performed, for example, the methods described may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation. Based on this understanding, the technical solution of the present application can be embodied in the form of a computer software product that is essentially or contributes to the existing technology. The computer software product is stored in a storage medium (such as ROM/RAM, disk , optical disk), including several instructions to cause a terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of this application.

The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above-mentioned specific implementations. The above-mentioned specific implementations are only illustrative and not restrictive. Those of ordinary skill in the art will Inspired by this application, many forms can be made without departing from the purpose of this application and the scope protected by the claims, all of which fall within the protection of this application.

Claims (13)

1. An image generation method applied to a first electronic device, the method comprising:

transmitting a first feature map corresponding to the first image to the second electronic equipment; the first image includes: a foreground image and a first background image; the first feature map is used for generating a second feature map by the second electronic device;

receiving the second feature map from the second electronic device, and generating a second background image according to the second feature map; the second background image and the first background image have different image contents;

a second image is generated based on the foreground image and the second background image.

2. The method of claim 1, wherein the foreground image comprises a person image;

Before the receiving the second feature map from the second electronic device, the method further includes:

determining a first mask image from the first image; the first mask image indicates a position of the person image in the first image;

transmitting the first mask image to the second electronic device; the first mask image is used for the second electronic device to determine a first characteristic area corresponding to the foreground image in the first characteristic image;

the second feature map is generated by the second electronic device according to a second feature region corresponding to the first background image in the first feature map, and the second feature region is a feature region except the first feature region in the first feature map.

3. The method of claim 2, wherein after the transmitting the first mask image to the second electronic device, the method further comprises:

receiving a second mask image from the second electronic device; the second mask image indicating a position of the foreground image in the first image, the second mask image being determined by the second electronic device based on the first mask image;

The generating a second image based on the foreground image and the second background image includes:

and fusing the foreground image and the second background image based on the second mask image to obtain the second image.

4. The method of claim 1, wherein before the sending the first feature map corresponding to the first image to the second electronic device, the method further comprises:

inputting the first image into an encoder to obtain the first characteristic diagram output by the encoder; the first feature map is obtained by encoding the first image by the encoder.

5. The method of claim 1, wherein the generating a second background image from the second feature map comprises:

inputting the second feature map into a decoder of the first electronic device to obtain the second background image output by the decoder; the second background image is obtained by decoding the second feature image by the decoder.

6. An image generation method applied to a second electronic device, the method comprising:

receiving a first feature map corresponding to a first image from first electronic equipment; the first image includes: a foreground image and a first background image;

Inputting the first feature map to a first model to obtain a second feature map output by the first model; the second feature map is generated by processing the first feature map by the first model;

transmitting the second feature map to the first electronic device; the second feature map is used for generating a second background image by the first electronic device, and generating a second image based on the foreground image and the second background image; the second background image and the first background image have different image contents.

7. The method of claim 6, wherein the foreground image comprises a person image;

before the first feature map is input to a first model to obtain a second feature map output by the first model, the method further includes:

receiving a first mask image from the first electronic device; the first mask image indicates a position of the person image in the first image;

determining a first feature area corresponding to the foreground image in the first feature map based on the first mask image;

the step of inputting the first feature map to a first model to obtain a second feature map output by the first model comprises the following steps:

Acquiring a second characteristic region except the first characteristic region from the first characteristic map, wherein the second characteristic region is a characteristic region corresponding to the first background image in the first characteristic map;

and inputting the second characteristic region into the first model to obtain the second characteristic diagram output by the first model.

8. The method of claim 7, wherein the foreground image further comprises a foreground image;

the determining, based on the first mask image, a first feature region corresponding to the foreground image in the first feature map includes:

determining a third characteristic region corresponding to the character image in the first characteristic image from the first characteristic image according to the first mask image;

and determining a fourth characteristic region corresponding to the third characteristic region and the foreground image in the first characteristic map as the first characteristic region.

9. The method of claim 7, wherein the foreground image further comprises a foreground image;

after the receiving a first mask image from the first electronic device, the method further comprises:

Determining a third mask image according to a fourth characteristic region corresponding to the foreground object image in the first characteristic map; the third mask image indicates a position of the foreground object image in the first image;

generating a second mask image from the first mask image and the third mask image; the second mask image indicates a position of the foreground image in the first image;

transmitting the second mask image to the first electronic device; the second mask image is used for fusing the foreground image and the second background image by the first electronic device to obtain the second image.

10. An image generating apparatus, the image generating apparatus being a first image generating apparatus, characterized in that the first image generating apparatus includes: the device comprises a sending module, a receiving module and a processing module;

the sending module is used for sending a first feature map corresponding to the first image to the second image generating device; the first image includes: a foreground image and a first background image; the first feature map is used for generating a second feature map by the second image generating device;

the receiving module is used for receiving the second characteristic diagram from the second image generating device;

The processing module is used for generating a second background image according to the second feature image received by the receiving module; the second background image and the first background image have different image contents; and generating a second image based on the foreground image and the second background image.

11. An image generating apparatus, which is a second image generating apparatus, characterized in that the second image generating apparatus includes: the device comprises a receiving module, a processing module and a sending module;

the receiving module is used for receiving a first feature map corresponding to the first image from the first image generating device; the first image includes: a foreground image and a first background image;

the processing module is used for inputting the first feature map received by the receiving module into a first model to obtain a second feature map output by the first model; the second feature map is generated by processing the first feature map by the first model;

the sending module is used for sending the second feature map generated by the processing module to the first image generating device; the second feature map is used for generating a second background image by the first image generating device and generating a second image based on the foreground image and the second background image; the second background image and the first background image have different image contents.

12. An electronic device comprising a processor and a memory storing a program or instructions executable on the processor, which when executed by the processor, implement the steps of the image generation method of any one of claims 1 to 9.

13. A readable storage medium, characterized in that the readable storage medium has stored thereon a program or instructions which, when executed by a processor, implement the steps of the image generation method according to any of claims 1 to 9.