WO2020093798A1

WO2020093798A1 - Method and apparatus for displaying target image, terminal, and storage medium

Info

Publication number: WO2020093798A1
Application number: PCT/CN2019/107085
Authority: WO
Inventors: 刘莹; 杨浩; 辛光
Original assignee: 北京达佳互联信息技术有限公司
Priority date: 2018-11-09
Filing date: 2019-09-20
Publication date: 2020-05-14
Also published as: CN109523461A

Abstract

The present application relates to the field of image processing, and relates to a method and apparatus for displaying a target image, a terminal, and a storage medium. The method comprises: obtaining a first face image and a second face image; generating a first facial topology image according to the first face image and a second facial topology image according to the second face image; fusing the first facial topological image and the second facial topological image to generate a target image; and displaying the target image on a target interface. By adopting the present application, the simulation effect may be better.

Description

Method, device, terminal and storage medium for displaying target image

This application requires the priority of the Chinese patent application filed on November 9, 2018 with the application number 201811334358.2 and the invention titled "Method, Device, Terminal and Storage Media for Displaying Target Images", the entire content of which is cited by reference Incorporated in this application.

Technical field

The present application relates to the field of image processing, and in particular, to a method, device, terminal, and storage medium for displaying a target image.

Background technique

At present, many users are very interested in simulating the appearance of the future baby. For example, simulating the image of the future baby through the images of two people is very interesting and a popular way of entertainment.

In the related art, when simulating a baby image through a parent ’s image, the user needs to first upload the father ’s image and the mother ’s image to the terminal. After receiving the two images, the terminal randomly selects one of the two images, and then Get pre-stored baby images in the image library. Match the selected image with each baby image separately to obtain the similarity between the selected image and each baby image, and then use the baby image with the highest similarity as the result of the simulation on the terminal's display screen display.

However, the inventor realized that the baby images simulated by the above processing are only the most similar to the parents ’images with respect to each baby image in the image library, and may not necessarily be very similar to the parents’ images, that is, according to The appearance of the baby obtained by the simulation of the parents' appearance may not be similar to that of the parents. In this way, the actual simulation result is different from the theoretically available simulation result, which makes the simulation effect poor.

Summary of the invention

The present application provides a method, device, terminal and storage medium for displaying a target image, which can solve the problem of poor simulation effect.

According to a first aspect of the embodiments of the present application, a method for displaying a target image is provided, including:

Obtain the first facial image and the second facial image;

Generating a first facial topology image based on the first facial image, and generating a second facial topology image based on the second facial image;

Fuse the first facial topology image with the second facial topology image to generate a target image;

The target image is displayed on the target interface.

According to a second aspect of the embodiments of the present application, there is provided an apparatus for displaying a target image, including:

An obtaining unit configured to obtain a first facial image and a second facial image;

A generating unit configured to generate a first facial topology image based on the first facial image, and generate a second facial topology image based on the second facial image;

The generating unit is further configured to fuse the first facial topology image and the second facial topology image to generate a target image;

The display unit is configured to display the target image on the target interface.

According to a third aspect of the embodiments of the present application, a terminal is provided, including:

processor;

Memory for storing processor executable instructions;

Wherein, the processor is configured to perform a method of displaying the target image.

According to a fourth aspect of the embodiments of the present application, there is provided a non-transitory computer-readable storage medium. When instructions in the storage medium are executed by a processor of a server, the server is enabled to perform a method of displaying a target image.

According to a fifth aspect of the embodiments of the present application, there is provided an application program, which causes the terminal to execute a method of displaying a target image when the application program is running on the terminal.

The technical solution provided by the embodiments of the present application may include the following beneficial effects: the terminal acquires the first facial image and the second facial image, and then generates the target image according to the fusion of the first facial image and the second facial image. In the related art, the baby image with the highest similarity to the parent's image is selected as the target image from the pre-stored baby images. The target image selected by this method is only the relatively similarity selected from the limited baby images. The largest image, but the selected image does not integrate the characteristics of the parent's image, so the similarity between the target image and the parent's image is low, and the simulation effect is poor. In this application, the generated target image merges the features of the first facial image and the second facial image. Therefore, compared with the related art, the target image generated by the application and the first facial image and the second facial image The similarity is greater, so the simulation effect is better.

BRIEF DESCRIPTION

The drawings herein are incorporated into the specification and constitute a part of the specification, show embodiments consistent with the application, and are used together with the specification to explain the principles of the application.

Fig. 1 is a flow chart of a method for displaying a target image according to an exemplary embodiment.

Fig. 2 is a flow chart of a method for displaying a target image according to an exemplary embodiment.

Fig. 3 is a schematic diagram showing an interface for displaying a target image according to an exemplary embodiment.

Fig. 4 is a schematic diagram showing a scene of displaying a target image according to an exemplary embodiment.

Fig. 5 is a schematic diagram showing an interface for displaying a target image according to an exemplary embodiment.

Fig. 6 is a block diagram of a device for displaying a target image according to an exemplary embodiment.

Fig. 7 is a schematic block diagram showing a structure of a terminal according to an exemplary embodiment;

Fig. 8 is a schematic block diagram of a specific structure of a terminal 700 according to an exemplary embodiment.

detailed description

Fig. 1 is a flowchart of a method for displaying a target image according to an exemplary embodiment. As shown in Fig. 1, the method is used in a terminal and includes the following steps.

In step 101, a first facial image and a second facial image are acquired.

In step 102, a first facial topology image is generated based on the first facial image, and a second facial topology image is generated based on the second facial image.

In step 103, the first facial topology image and the second facial topology image are fused to generate a target image.

In step 104, a target image is displayed on the target interface.

Optionally, fusing the first facial topology image with the second facial topology image to generate the target image includes:

A plurality of first region sub-images in the first facial topology image and a plurality of corresponding second region sub-images in the second facial topology image are respectively fused to generate a target image.

Optionally, fusing a plurality of first region sub-images in the first facial topology image and a plurality of corresponding second-region sub images in the second facial topology image to generate a target image, including:

Adjusting the shape and size of the corresponding plurality of second area sub-images in the second facial topology image according to the plurality of first area sub-images in the first facial topology image;

Fit the adjusted multiple second-region sub-images to the corresponding multiple first-region sub-images, and according to the pre-stored pixel fusion algorithm, the first-region sub-image and the second face in the first facial topology image The corresponding sub-images of the second region in the topological image are fused to generate the target image.

Optionally, fusing the first region sub-image in the first facial topology image with the corresponding second region sub-image in the second facial topology image to generate the target image includes:

Adjust the first facial topology image according to the first weight corresponding to the first facial topology image to obtain the adjusted first facial topology image;

Adjust the second facial topology image according to the second weight corresponding to the second facial topology image to obtain the adjusted second facial topology image;

The first region sub-image in the adjusted first facial topology image is fused with the corresponding second region sub-image in the adjusted second facial topology image to generate a target image.

Optionally, the first weight is determined randomly, and the second weight is the difference between 1 and the first weight.

Optionally, after generating the target image, the method further includes:

Acquiring a first skin color value at a preset position of the first facial topology image, and acquiring a second skin color value at a preset position of the second facial topology image;

Determine the target skin color value at the preset position of the target image according to the first skin color value and the second skin color value;

According to the target skin color value, the skin color value at the preset position in the target image is adjusted.

Optionally, according to the first skin color value and the second skin color value, determining the target skin color value at the preset position of the target image includes:

Obtain preset skin tone adjustment values;

The sum of the first skin color value and the second skin color value is obtained, and the difference obtained by subtracting the skin color adjustment value from the sum value is determined as the target skin color value at the preset position of the target facial topology image.

Optionally, acquiring the first facial image and the second facial image includes:

When receiving the target image generation instruction, the video frame is collected through the camera, and one frame of the video frame is randomly selected as the source image from the collected multi-frame video frames;

Extract the first facial image and the second facial image from the source image;

After acquiring the first facial image and the second facial image, the method further includes:

On the target interface, the first facial image and the second facial image are displayed.

Optionally, generating a first facial topology image based on the first facial image and generating a second facial topology image based on the second facial image include:

In the first facial image, determine the first set of facial feature points;

Connect the facial feature points in the first group of facial feature points according to the preset connection rule to generate the first facial topology image;

In the second facial image, determine the second set of facial feature points;

Connect the facial feature points in the second group of facial feature points according to the preset connection rule to generate the second facial topology image.

Optionally, the method further includes:

Obtain an update patch package, in which the update patch package carries instructions for calling the CPU function of the central processor and instructions for calling the GPU function of the image processor. The CPU functions include using sensors, detecting touch events, detecting trigger events, and determining facial feature points. ;

Load the update patch package.

The embodiments of the present application will introduce a method of displaying a target image in combination with a specific implementation manner. This method may be implemented by a terminal, which may be a terminal installed with an application program that displays a target image. The terminal includes at least a graphics card and a CPU (Central Processing Unit, central processing unit). As shown in the flowchart of the method for generating a target image shown in FIG. 2, in this embodiment, the first facial image and the second facial image are the father's facial image and the mother's facial image, respectively, and the generated target image is the baby image. The method for generating the target image will be described. The processing flow of this method may include the following steps:

In step 201, when the terminal receives the target image generation instruction, it collects video frames through the camera, randomly selects one frame of video frames from the collected multi-frame video frames as the source image, and then extracts the first The facial image and the second facial image, and then, on the target interface of the terminal, the first facial image and the second facial image are displayed.

The source image refers to an image containing at least two facial images to be fused.

When the user wants to generate a baby image, that is, when the user wants to generate a target image, the user can first open the application installed on the terminal, then, the user clicks the option to generate a baby image, the terminal receives the option to generate a baby image The instruction is the target image generation instruction. Then, the terminal starts the camera function and collects video frames through the camera. When the terminal receives the video capture stop instruction, or when the duration of capturing video frames reaches the preset duration, the terminal stops capturing video frames and randomly selects at least one video frame among the multi-frame video frames that have been captured As the source image.

Then in the source image, the terminal can use the pre-stored face recognition algorithm to identify two face images in the obtained source image to obtain the first face image and the second face image. Then, the terminal on the target interface The first facial image and the second facial image are displayed. Among them, the face recognition algorithm can be a local feature analysis method, a feature face method, a recognition algorithm based on an elastic model, a neural network recognition algorithm, a hidden Markov model algorithm, etc., as long as the first facial image can be extracted from the source image As well as the algorithm of the second facial image, this application does not limit it.

If the terminal recognizes more than two facial images in the source image, the terminal can intercept the multiple recognized facial images and display them to the user, so that the user can manually select two of the facial images as the first facial image and the first Two face images, as shown in Figure 3. Alternatively, the terminal may also send an image error prompt message to the user to prompt the user to upload the source image again, which is not limited in this application. If the terminal cannot recognize two facial images in the source image, it will send an image error prompt message to prompt the user to upload the source image again. There are many ways to send an image error prompt message, such as displaying prompt text, displaying a prompt picture, and issuing a voice prompt. As long as it can play the role of prompting the user to re-upload the source image, this application will send an image error prompt message The method is not limited.

The terminal determines the processing method of the source image. In addition to the processing methods provided in the above steps, other processing methods may also be used. For example, the terminal may receive at least one image uploaded by the user as a source image. In this way, the user can choose to upload the image stored in the terminal in advance, or can choose to take the image and upload the taken image. Then, the user can upload at least one image, and the terminal uses the received image as the source image. For another example, the terminal may receive at least one image and video frame uploaded by the user as the source image. In this way, the user can upload at least one image and a piece of video. The terminal selects at least one video frame from the received video, and uses the selected video frame and the received image as the source image. The terminal determines the source image in various processing methods, and can select a corresponding processing method according to actual needs, which is not limited in this application.

In step 202, the terminal generates a first facial topology image based on the first facial image, and generates a second facial topology image based on the second facial image.

In a feasible embodiment, the step of generating a facial topology image based on the facial image in the above step 202 may be as follows: In the facial image, determine a set of facial feature points; according to the preset connection rule, the set of facial features The facial feature points in the points are connected to generate a facial topology image, as shown in Figure 4. Taking the first facial image as an example, the above steps may be: in the first facial image, determine the first set of facial feature points; according to the preset connection rule, the facial features in the first set of facial feature points The points are connected to generate the first facial topology image. Correspondingly, the above step 202 includes the following steps 2021 and 2022.

In step 2021, after determining the first facial image and the second facial image, the terminal may use a pre-stored facial feature point recognition algorithm to calibrate a set of facial feature points in the first facial image (ie, the first set of faces Feature points). Based on the same algorithm, the second set of facial feature points are calibrated in the second facial image. Among them, facial feature point recognition algorithms can be CLM (Constrained local model, constrained local model) algorithm, Cascaded Regression (cascade regression, a facial feature point positioning method) algorithm, CNN (Convolutional Neural Network, convolutional neural network) Algorithms, etc., as long as they can implement algorithms for calibrating facial feature points in facial images, this application does not limit this.

For any facial image, the same facial feature point recognition algorithm is used to calibrate the facial image, the number of calibrated facial feature points is the same, and each facial feature point has an identifier, and each identifier corresponds to the facial feature point It has a fixed meaning, for example, the facial feature points of logos 19 to 24 represent the upper edge of the left eyebrow in the facial image, and the facial feature points of logo 49 represent the left eye corner of the left eye in the facial image.

In step 2022, the terminal acquires a preset connection rule, connects facial feature points, and generates a first facial topology image of the first facial image. The preset connection rule may be the identification connection rule of the facial feature points. In this case, the facial connection points specified in the identification connection rule to be connected are connected through the predetermined identification connection rule, and the first face The image is divided into multiple areas, and each area can be called an area sub-image, and there is no overlap between all area sub-images. Generally speaking, the logo connection rule stipulates that certain three facial feature points are connected, that is, the divided sub-images are triangular images. Of course, other numbers of facial feature points can be specified in the logo connection rule according to actual needs. Connected, that is, the divided sub-images are images of other shapes, which is not limited in this application.

Similarly, the processing method for generating the second facial topological image may be: in the second facial image, determine the second group of facial feature points; according to the preset connection rule, the facial feature points in the second group of facial feature points Connect to generate a second facial topology image. The corresponding processing method for generating the second facial topology image can refer to the above processing method for generating the first facial topology image, which will not be repeated here.

In step 203, the terminal adjusts the first facial topology image according to the first weight corresponding to the first facial topology image to obtain the adjusted first facial topology image, and adjusts the second facial topology image according to the second weight To obtain the adjusted second facial topology image.

The terminal first obtains the first weight corresponding to the first facial topology image and the second weight corresponding to the second facial topology image, where the first weight may be the weight of the entire first facial topology image. In this case, the first The second weight is the weight of the entire second facial topology image. In addition, the first weight may also be the weight of a certain area sub-image in the first face topology image. In this case, the second weight is the area sub-image corresponding to the first weight in the second face topology image The weight of the region sub-image corresponding to the image, the first weight and the second weight may each include a plurality of different weight values. In addition, the first weight may also be the weight of the facial feature image composed of multiple region sub-images in the first facial topology image, and the second weight is the weight of the corresponding facial feature image in the second facial topology image. Both the first weight and the second weight may include a plurality of different weight values. For example, if the first weight is the weight of the eyes in the first facial topology image, the second weight is the weight of the eyes in the second facial topology image. For the above method, this application does not specifically limit this.

There may be many ways to obtain the first weight and the second weight, and several methods are listed below.

Method 1: The technician can preset the value of the first weight and the value of the second weight. When the terminal fuses the first facial topology image and the second facial topology image, the first weight and the second weight correspond to The storage area may directly read the value of the first weight and the value of the second weight.

Method 2: The user can determine the value of the first weight and the value of the second weight. For example, if the terminal provides the user with an option to adjust the similarity between the generated baby image and the two facial images, the user can manually adjust the similarity. The degree of similarity is converted into the value of the first weight and the value of the second weight.

Method 3: The terminal can randomly set the value of the first weight, then the second weight is 1 minus the difference of the first weight, so that the user does not know which facial image the generated baby image will be more similar to, which increases The uncertainty of the generated baby image also increases the fun.

After acquiring the first weight and the second weight through the above processing steps, the terminal adjusts the first facial topology image according to the first weight to obtain the adjusted first facial topology image, and adjusts the second face according to the second weight Topology image to obtain the adjusted second face topology image.

For the case where the first weight is the weight of the entire first facial topology image and the second weight is the weight of the entire second facial topology image, for the method 1, method 2, and method 3 in step 203 above, the processing in this step is: : Determine the RGB (Red-Green-Blue, red-green-blue) value of each pixel of the first facial topology image, multiply the RGB value of each pixel by the first weight, and the obtained product is the adjusted first The RGB value of a facial topology image. According to the adjusted RGB value of the first facial topology image, the RGB value of the first facial topology image is adjusted, that is, the RGB value of the first facial topology image is set to the adjusted RGB value. Similarly, according to the second weight, the second facial topology image is adjusted to obtain the adjusted second facial topology image. For the processing steps of obtaining the adjusted second facial topology image, refer to the above-mentioned generating the adjusted first facial topology image, which will not be repeated here.

For the case where the first weight is the weight of a certain area sub-image in the first face topology image, and the second weight is the weight of the area sub-image corresponding to the area sub-image corresponding to the first weight in the second face topology image For the method 1, method 2, and method 3 in the above step 203, the processing method of this step is: taking an area sub-image in the first facial topology image as an example, determine each pixel in the area sub-image RGB value, multiply each RGB value with the first weight, the product is the adjusted RGB value of the sub-image of the area, adjust the sub-image of the area according to the RGB value, that is to say, the The RGB value of the area sub-image is set to this RGB value. Based on the same processing method, after adjusting each region sub-image in the first facial topology image, the adjusted first facial topology image is obtained. Similarly, according to the second weight, the second facial topology image is adjusted to obtain the adjusted second facial topology image. For the processing steps of obtaining the adjusted second facial topology image, refer to the above-mentioned generating the adjusted first facial topology image, which will not be repeated here.

For the case where the first weight is the weight of a facial feature image composed of a plurality of regional sub-images in the first facial topology image, and the second weight is the weight of the corresponding facial feature image in the second facial topology image, for the above method 1, Method two and method three, the processing method of this step is: taking a certain facial image composed of multiple area sub-images in the first facial topology image as an example, determine the RGB value of each pixel in the facial image, and Each RGB value is multiplied by the first weight, and the obtained product is the adjusted RGB value of the facial image, and the facial image is adjusted according to the RGB value, that is, the RGB value of the facial image is set to The RGB value. Based on the same processing method, after adjusting each facial feature image in the first facial topology image, an adjusted first facial topology image is obtained. Similarly, according to the second weight, the second facial topology image is adjusted to obtain the adjusted second facial topology image. For the processing steps of obtaining the adjusted second facial topology image, refer to the above-mentioned generating the adjusted first facial topology image, which will not be repeated here.

In step 204, the terminal fuses the first region sub-image in the adjusted first facial topology image with the corresponding second region sub-image in the adjusted second facial topology image to generate a target image.

Among them, the first region sub-images are connected according to preset facial feature points, and a part of the images divided in the first face topology image, similarly, the second region sub-images are connected according to preset facial feature points , A part of the image divided in the second facial topology image. The correspondence between the first region sub-image and the second region sub-image means that the identifier of the corresponding facial feature point of the first region sub-image is the same as the identifier of the corresponding facial feature point of the second region sub-image.

After adjusting the first face topology image and the second face topology image according to the first weight and the second weight through the above steps, the terminal will adjust the adjusted first face topology image according to a preset image fusion algorithm The sub-image of the first region of the image is fused with the sub-image of the second region corresponding to the topological image of the adjusted second face to generate a target image. Among them, there can be many image fusion algorithms, such as pixel-level image fusion algorithm, Poisson fusion algorithm or MVC (Mean Values, Mean Coordinates) image fusion algorithm, etc., as long as the first facial topology image and the second facial Any algorithm for topological image fusion may be used, which is not limited in this application.

The following uses the image fusion algorithm as a pixel-level image fusion algorithm for illustration. The above step 204 may include the following steps 2041-2042.

In step 2041, the terminal adjusts the shape and size of the corresponding multiple second-region sub-images in the second facial topology image according to the multiple first-region sub-images in the first facial topology image.

The terminal pre-sets a facial topology image as a base image, and another facial topology image as a material, assuming that the terminal determines the first facial topology image as a base image, taking a second region sub-image in the second facial topology image as an example , The terminal first determines the first area sub-image corresponding to the second area sub-image in the first facial topology image, and then, based on the determined shape and size of the first area sub-image, the terminal The shape and size are adjusted so that the second region sub-image and the corresponding first region sub-image have the same shape and size.

In step 2042, the adjusted plurality of second region sub-images are attached to the corresponding plurality of first region sub-images, and according to the pre-stored pixel fusion algorithm, the first region sub-image in the first facial topology image The image is fused with the corresponding second region sub-image in the second facial topology image to generate the target image.

The above steps 203-204 are to adjust the weights of the first facial topology image and the second facial topology image, and then fuse the adjusted first facial topology image and the second facial topology image to generate the target image. Of course, The weights of the first facial topology image and the second facial topology image may not be adjusted, but after step 202, the first facial topology image and the second facial topology image are directly fused to generate a target image. No limitation.

Optionally, in order to make the generated baby image more beautiful and more in line with the baby's image, the target image can be correspondingly beautified, for example, the facial contour in the target image can be liquefied to make the facial contour more round, Enlarge the eye contour in the target image to make the eye contour bigger and more round, etc., or adjust the skin color value of the target image.

In step 205, a first skin color value is obtained at a preset position of the first facial topology image, and a second skin color value is obtained at a preset position of the second facial topology image.

Wherein, the preset position is the position of the preset area sub-image, and the skin color value is the RGB value of a pixel in the facial topology image.

In order to further reflect the association between the generated baby image and the parent ’s facial image, the skin color value of the baby image may be determined according to the skin color value of the parent ’s facial image, that is, the first facial topology image is extracted at the preset position Skin color value, the second skin color value is extracted at a preset position of the second facial topology image.

In step 206, the target skin color value at the preset position of the target image is determined according to the first skin color value and the second skin color value.

After determining the first skin color value and the second skin color value, the terminal may calculate the average value of the first skin color value and the second skin color value, and determine it as the target skin color value at the preset position of the target image.

In addition, considering that the skin color of the baby is generally white, when determining the target skin color value at the preset position according to the skin color value of the parent's facial image, the skin color value of the parent's facial image can be appropriately adjusted and correspondingly processed The method may be as follows: acquiring a preset skin tone adjustment value; acquiring the sum of the first skin tone value and the second skin tone value, and subtracting the difference obtained by subtracting the skin tone adjustment value from the sum value to determine the preset position of the target facial topology image Target skin color value.

In addition, the technician can also pre-set the target skin tone value at the preset position and store the target skin tone value in the terminal. When the terminal acquires the target skin tone value at the preset position, in the storage area corresponding to the target skin tone value Simply read the target skin color value.

In step 207, according to the target skin color value, the terminal adjusts the skin color value at the preset position in the target image.

When the terminal adjusts the skin color value at the preset position in the target image according to the target skin color value, the RGB value of each pixel at the preset position is adjusted to the target skin color value.

In step 208, the terminal displays the target image after adjusting the skin color value on the target interface.

After adjusting the skin color value at the preset position of the target image, the terminal displays the target image, as shown in FIG. 5. In addition, when the terminal displays the target image, in order to increase the beauty and interest, it can first play the pre-stored audio, animation effects, etc., and then display the target image, or display the target image according to the preset display special effects, according to the actual needs The corresponding display settings can be made, which is not limited in this application.

In the related art, when the application program is updated, a new version of the installation package may need to be installed on the terminal to add new functions. However, this wastes network resources, and technicians need to make a new version of the installation package when updating, which wastes manpower and material resources and cannot be updated quickly. In order to achieve a rapid update iteration, the technical personnel can write the server's graphics card driver script to the terminal in the form of a string when writing the update patch package of the above function. When the user obtains the above functions through updating, the terminal obtains an update patch package, in which the update patch package carries instructions for calling the CPU function of the central processor and instructions for calling the GPU function of the image processor. The CPU functions include using sensors and detecting touch Events, detect trigger events, determine facial feature points. Technicians encapsulate high-latitude CPU logic by using dynamic scripting languages, that is, encapsulate Opengl ES-related primitives into instructions that dynamic scripts can call, and the client-side sensors, touch events, trigger events, and facial key points Recognition functions and other high-level logic are encapsulated in the script driver. Then, the terminal loads the update patch package, that is, it can be updated without issuing a new version of the installation package, which accelerates the iteration efficiency.

In the method provided in the embodiment of the present application, the terminal acquires the first facial image and the second facial image, and then generates the first facial topology image and the second facial topology image according to the first facial image and the second facial image, respectively, and then, Adjust the first facial topology image and the second facial topology image according to the first weight and the second weight respectively, and then fuse the adjusted first facial topology image with the adjusted second facial topology image to generate the target image . In this way, the generated target image merges the features of the first facial image and the second facial image. Therefore, the generated target image has a large similarity with the first facial image and the second facial image, making the simulation effect better. In addition, the generated target image can be adjusted according to the target skin color value, so that the skin color of the generated target image is more in line with the characteristics of the baby, and the generated target image is more beautiful.

Fig. 6 is a block diagram of a device for displaying a target image according to an exemplary embodiment. 6, the device includes an acquisition unit 610, a generation unit 620, and a display unit 630.

The obtaining unit 610 is configured to obtain a first facial image and a second facial image;

The generating unit 620 is configured to generate a first facial topology image based on the first facial image and generate a second facial topology image based on the second facial image;

The generating unit 620 is further configured to fuse the first facial topology image and the second facial topology image to generate a target image;

The display unit 630 is configured to display the target image on the target interface.

Optionally, the generating unit 620 is configured to:

Optionally, the acquiring unit 610 is further configured to acquire the first skin color value at a preset position of the first facial topology image and acquire the second at the preset position of the second facial topology image after generating the target image Skin color value

The device also includes:

The determining unit is configured to determine the target skin color value at the preset position of the target image according to the first skin color value and the second skin color value;

The adjusting unit is configured to adjust the skin color value at the preset position in the target image according to the target skin color value.

Optionally, the determination unit is configured to:

Obtain preset skin tone adjustment values;

Optionally, the obtaining unit 610 is configured to:

The display unit 630 is configured to display the first facial image and the second facial image on the target interface after acquiring the first facial image and the second facial image.

Optionally, the generating unit 620 is configured to:

In the first facial image, determine the first set of facial feature points;

In the second facial image, determine the second set of facial feature points;

Optionally, the acquiring unit 610 is configured to acquire an update patch package, wherein the update patch package carries instructions for calling the CPU function of the central processor and instructions for calling the GPU function of the image processor. The CPU functions include using sensors, Functions to detect touch events, detect trigger events, and determine facial feature points;

The device also includes:

The loading unit is configured to load the update patch package.

Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment related to the method, and will not be elaborated here.

Fig. 7 is a schematic structural diagram of a terminal according to an exemplary embodiment, including: a processor 71; a memory 72 for storing processor executable instructions; wherein, the processor 71 is configured to execute any of the above The method for displaying the target image described in the embodiment.

Fig. 8 is a specific structural block diagram of a terminal 700 according to an exemplary embodiment. For example, the terminal 700 may be a mobile phone, a computer, a messaging device, a game console, a tablet device, or the like.

8, the terminal 700 may include one or more of the following components: a processing component 702, a memory 704, a power component 706, a multimedia component 708, an audio component 710, an input / output (I / O) interface 712, a sensor component 714,与通信组 716.

The processing component 702 generally controls the overall operations of the device 700, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 702 may include one or more processors 720 to execute instructions to complete all or part of the steps in the above method. In addition, the processing component 702 may include one or more modules to facilitate interaction between the processing component 702 and other components. For example, the processing component 702 may include a multimedia module to facilitate interaction between the multimedia component 708 and the processing component 702.

The memory 704 is configured to store various types of data to support operations at the terminal 700. Examples of these data include instructions for any application or method operated on the terminal 700, contact data, phone book data, messages, pictures, videos, and the like. The memory 704 may be implemented by any type of volatile or nonvolatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable and removable Programmable read only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

The power supply component 706 provides power to various components of the terminal 700. The power component 706 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the device 700.

The multimedia component 708 includes a screen that provides an output interface between the terminal 700 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, swipe, and gestures on the touch panel. The touch sensor may not only sense the boundary of the touch or sliding action, but also detect the duration and pressure related to the touch or sliding operation. In some embodiments, the multimedia component 708 includes a front camera and / or a rear camera. When the terminal 700 is in an operation mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have focal length and optical zoom capabilities.

The audio component 710 is configured to output and / or input audio signals. For example, the audio component 710 includes a microphone (MIC). When the terminal 700 is in an operation mode, such as a call mode, a recording mode, and a voice recognition mode, the microphone is configured to receive an external audio signal. The received audio signal may be further stored in the memory 704 or sent via the communication component 716. In some embodiments, the audio component 710 further includes a speaker for outputting audio signals.

The I / O interface 712 provides an interface between the processing component 702 and a peripheral interface module. The peripheral interface module may be a keyboard, a click wheel, or a button. These buttons may include, but are not limited to: home button, volume button, start button, and lock button.

The sensor component 714 includes one or more sensors for providing the terminal 700 with status evaluation in various aspects. For example, the sensor component 714 can detect the on / off state of the device 700 and the relative positioning of the components, for example, the components are the display and keypad of the device 700, and the sensor component 714 can also detect the position change of the terminal 700 or a component of the terminal 700 The presence or absence of user contact with the terminal 700, the orientation or acceleration / deceleration of the terminal 700, and the temperature change of the terminal 700. The sensor assembly 714 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor component 714 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor component 714 may further include an acceleration sensor, a gyro sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 716 is configured to facilitate wired or wireless communication between the terminal 700 and other devices. The terminal 700 may access a wireless network based on a communication standard, such as WiFi, an operator network (such as 2G, 3G, 4G, or 5G), or a combination thereof. In an exemplary embodiment, the communication component 716 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 716 further includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, the terminal 700 may be one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable A gate array (FPGA), controller, microcontroller, microprocessor or other electronic components are implemented to perform the above method.

In an exemplary embodiment, there is also provided a computer-readable storage medium including instructions, for example, a memory 704 including instructions, the above instructions may be executed by the processor 720 of the terminal 700 to complete the above method of displaying a target image, the method includes : Acquire the first facial image and the second facial image; generate the first facial topology image based on the first facial image, generate the second facial topology image based on the second facial image; combine the first facial topology image with the second face The topological images are fused to generate the target image; the target image is displayed on the target interface. For example, the non-transitory computer-readable storage medium may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, or the like.

In an exemplary embodiment, an application program is also provided, which includes one or more instructions, and the one or more instructions may be executed by the processor 720 of the terminal 700 to complete the above method of displaying a target image, the method includes : Acquire a first facial image and a second facial image; generate a first facial topology image based on the first facial image, generate a second facial topology image based on the second facial image; compare the first facial topology image with the first The two facial topological images are fused to generate a target image; the target image is displayed on the target interface. Optionally, the above instructions may also be executed by the processor 720 of the terminal 700 to complete other steps involved in the above exemplary embodiments.

Claims

A method for displaying a target image includes:

Obtain the first facial image and the second facial image;

Generating a first facial topology image based on the first facial image, and generating a second facial topology image based on the second facial image;

Fuse the first facial topology image with the second facial topology image to generate a target image;

The target image is displayed on the target interface.
The method of displaying a target image according to claim 1, the fusing the first facial topology image and the second facial topology image to generate a target image includes:

A plurality of first area sub-images in the first face topology image and a plurality of corresponding second area sub-images in the second face topology image are respectively fused to generate a target image.
The method for displaying a target image according to claim 2, wherein the plurality of first area sub-images in the first facial topology image and the corresponding second area sub-images in the second facial topology image The images are fused separately to generate the target image, including:

Adjusting the shape and size of the corresponding second area sub-images in the second facial topology image according to the multiple first area sub-images in the first facial topology image;

The adjusted plurality of second region sub-images are attached to the corresponding plurality of first region sub-images, and according to the pre-stored pixel fusion algorithm, the first region sub-image in the first facial topology image is The sub-images corresponding to the second region in the second facial topology image are fused to generate a target image.
The method for displaying a target image according to claim 2, wherein the plurality of first area sub-images in the first facial topology image and the corresponding second area sub-images in the second facial topology image The images are fused separately to generate the target image, including:

Adjusting the first facial topology image according to the first weight corresponding to the first facial topology image to obtain the adjusted first facial topology image;

Adjusting the second facial topology image according to the second weight corresponding to the second facial topology image to obtain the adjusted second facial topology image;

The first region sub-image in the adjusted first facial topology image is fused with the corresponding second region sub-image in the adjusted second facial topology image to generate a target image.
The method for displaying a target image according to claim 4, wherein the first weight is determined randomly, and the second weight is a difference between 1 and the first weight.
The method for displaying a target image according to claim 1, after generating the target image, further comprising:

Acquiring a first skin color value at a preset position of the first facial topology image, and acquiring a second skin color value at a preset position of the second facial topology image;

Determine the target skin color value at the preset position of the target image according to the first skin color value and the second skin color value;

According to the target skin color value, the skin color value at a preset position in the target image is adjusted.
The method for displaying a target image according to claim 6, the determining the target skin color value at a preset position of the target image according to the first skin color value and the second skin color value includes:

Obtain preset skin tone adjustment values;

Obtaining the sum of the first skin color value and the second skin color value, and subtracting the difference between the sum value and the skin color adjustment value to determine the target skin color at the preset position of the target facial topology image value.
The method for displaying a target image according to claim 1, the acquiring the first facial image and the second facial image includes:

When receiving the target image generation instruction, the video frame is collected through the camera, and one frame of the video frame is randomly selected as the source image from the collected multi-frame video frames;

Extracting the first facial image and the second facial image from the source image;

After acquiring the first facial image and the second facial image, the method further includes:

On the target interface, the first facial image and the second facial image are displayed.
The method of displaying a target image according to claim 1, said generating a first facial topology image based on the first facial image and generating a second facial topology image based on the second facial image includes:

In the first facial image, determine a first set of facial feature points;

Connect each facial feature point in the first group of facial feature points according to a preset connection rule to generate a first facial topology image;

In the second facial image, determine a second set of facial feature points;

Connect the facial feature points in the second group of facial feature points according to a preset connection rule to generate a second facial topology image.
The method of displaying a target image according to claim 1, further comprising:

Obtain an update patch package, where the update patch package carries instructions to call the CPU function of the central processor and instructions to call the GPU function of the image processor. The CPU functions include using sensors, detecting touch events, detecting trigger events, and determining Facial feature point function;

Load the update patch package.
A device for displaying target images, including:

An obtaining unit configured to obtain a first facial image and a second facial image;

A generating unit configured to generate a first facial topology image based on the first facial image, and generate a second facial topology image based on the second facial image;

The generating unit is further configured to fuse the first facial topology image and the second facial topology image to generate a target image;

The display unit is configured to display the target image on the target interface.
The apparatus for displaying a target image according to claim 11, wherein the generating unit is configured to:

A plurality of first area sub-images in the first face topology image and a plurality of corresponding second area sub-images in the second face topology image are respectively fused to generate a target image.
The apparatus for displaying a target image according to claim 12, wherein the generating unit is configured to:

Adjusting the shape and size of the corresponding second area sub-images in the second facial topology image according to the multiple first area sub-images in the first facial topology image;

The adjusted plurality of second region sub-images are attached to the corresponding plurality of first region sub-images, and according to the pre-stored pixel fusion algorithm, the first region sub-image in the first facial topology image is The sub-images corresponding to the second region in the second facial topology image are fused to generate a target image.
The apparatus for displaying a target image according to claim 12, wherein the generating unit is configured to:

Adjusting the first facial topology image according to the first weight corresponding to the first facial topology image to obtain the adjusted first facial topology image;

Adjusting the second facial topology image according to the second weight corresponding to the second facial topology image to obtain the adjusted second facial topology image;

The first region sub-image in the adjusted first facial topology image is fused with the corresponding second region sub-image in the adjusted second facial topology image to generate a target image.
The apparatus for displaying a target image according to claim 14, wherein the first weight is determined randomly, and the second weight is a difference between 1 and the first weight.
The apparatus for displaying a target image according to claim 11, further comprising:

The acquiring unit is further configured to acquire a first skin color value at a preset position of the first facial topology image and acquire a second at a preset position of the second facial topology image after generating the target image Skin color value

A determining unit configured to determine a target skin color value at a preset position of the target image according to the first skin color value and the second skin color value;

The adjusting unit is configured to adjust the skin color value at a preset position in the target image according to the target skin color value.
The apparatus for displaying a target image according to claim 16, the determination unit is configured to:

Obtain preset skin tone adjustment values;

Obtaining the sum of the first skin color value and the second skin color value, and subtracting the difference between the sum value and the skin color adjustment value to determine the target skin color at the preset position of the target facial topology image value.
The apparatus for displaying a target image according to claim 11, the acquisition unit is configured to:

When receiving the target image generation instruction, the video frame is collected through the camera, and one frame of the video frame is randomly selected as the source image from the collected multi-frame video frames;

Extracting the first facial image and the second facial image from the source image;

The display unit is configured to display the first facial image and the second facial image on the target interface after acquiring the first facial image and the second facial image.
The apparatus for displaying a target image according to claim 11, wherein the generating unit is configured to:

In the first facial image, determine a first set of facial feature points;

Connect each facial feature point in the first group of facial feature points according to a preset connection rule to generate a first facial topology image;

In the second facial image, determine a second set of facial feature points;

Connect the facial feature points in the second group of facial feature points according to a preset connection rule to generate a second facial topology image.
The apparatus for displaying a target image according to claim 11, the method further comprising:

The acquiring unit is configured to acquire an update patch package, wherein the update patch package carries instructions to call the CPU function of the central processor and instructions to call the GPU function of the image processor. The CPU functions include using a sensor and detecting Touch events, detect trigger events, determine facial feature points;

The loading unit is configured to load the update patch package.
A terminal, including:

processor;

Memory for storing processor executable instructions;

Wherein, the processor is configured to:

Obtain the first facial image and the second facial image;

Generating a first facial topology image based on the first facial image, and generating a second facial topology image based on the second facial image;

Fuse the first facial topology image with the second facial topology image to generate a target image;

The target image is displayed on the target interface.
A computer-readable storage medium, when instructions in the storage medium are executed by a processor of a server, enable the server to execute a method of displaying a target image, the method including:

Obtain the first facial image and the second facial image;

Generating a first facial topology image based on the first facial image, and generating a second facial topology image based on the second facial image;

Fuse the first facial topology image with the second facial topology image to generate a target image;

The target image is displayed on the target interface.