Image processing method, device and electronic equipment

Abstract

The present invention discloses an image processing method, device and electronic equipment, which includes: acquiring a first image including inner clothing, and a second image including a human body wearing outer clothing; performing deformation processing on the inner clothing in the first image according to the second image to obtain a deformed inner clothing that matches the posture of the human body; performing a first segmentation processing on the human body in the second image to obtain a segmentation result of the human body; obtaining a third image including the human body wearing the inner clothing and the outer clothing according to the segmentation result and the deformed inner clothing; and outputting the third image.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing technology, and more particularly, to an image processing method, an image processing device, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the development of the Internet, online shopping is gradually replacing physical purchases due to its advantages such as abundant resources and high efficiency. However, the fitting effect of clothing products is not ideal due to the lack of personal try-on and variety matching.

[0003] In the existing technology, the solutions available on the market can be divided into two categories: the first category uses generative adversarial networks to directly generate clothing in one image onto the model in another image; the second category uses 3D technology to "wear" the 3D model of the clothing on a 3D virtual model.

[0004] However, the disadvantage of the first type of solution is that it cannot accurately control the texture pattern of the clothing to remain unchanged during the generation process, and there is a high probability of confusion (for example, the brand logo is deformed or blurred). Therefore, the first type of solution is only suitable for clothing without complex texture patterns. In addition, when this solution handles clothing with complex patterns, the upper body effect is very poor (for example, the collar and sleeves do not correspond to the real position), and there are strict requirements for the target model image (for example, the model image must be a white background image). The disadvantage of the second type of solution is that the cost of 3D modeling technology is too high, and it is only suitable for small-scale production and cannot be promoted on a large scale. In addition, the realism of 3D clothing and 3D models is still far from that of real photos, which seriously reduces the display effect. Summary of the Invention

[0005] One purpose of the present disclosure is to provide a new technical solution for realizing virtual fitting.

[0006] According to a first aspect of the present disclosure, there is provided an image processing method, comprising:

[0007] Acquire a first image including inner clothing and a second image including a person wearing outer clothing;

[0008] performing deformation processing on the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body;

[0009] performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body;

[0010] obtaining, based on the segmentation result and the deformed inner garment, a third image of a human body wearing the inner garment and the outer garment;

[0011] The third image is output.

[0012] Optionally, deforming the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body includes:

[0013] performing a second segmentation process on the inner garment in the first image to obtain a front piece of the inner garment;

[0014] extracting key points of the inner garment in the first image;

[0015] Extracting key points of the human body in the second image;

[0016] The deformed inner garment is obtained according to the front piece of the inner garment, the human body in the second image, the key points of the inner garment and the key points of the human body.

[0017] Optionally, obtaining the deformed inner garment according to the front piece of the inner garment, the human body in the second image, the key points of the inner garment, and the key points of the human body includes:

[0018] Obtaining a first projection matrix based on the front piece of the inner garment and a preset reference human body, and obtaining a second projection matrix based on the human body in the second image and the reference human body;

[0019] Processing the front piece of the inner garment according to the first projection matrix to obtain an inner garment projection; processing the human body according to the second projection matrix to obtain a human body projection;

[0020] The deformed inner layer is obtained according to the inner layer projection and the human body projection, the inner layer key points and the human body key points.

[0021] Optionally, obtaining, based on the segmentation result and the deformed inner garment, a third image of a human body wearing the inner garment and the outer garment includes:

[0022] Fusing the segmentation result and the deformed inner layer to obtain a fused image;

[0023] The portion of the deformed inner garment that exceeds the body range of the human body in the fused image is erased to obtain the third image.

[0024] Optionally, the method further includes:

[0025] Perform skin repair processing on the inner layer area in the segmentation result.

[0026] Optionally, fusing the segmentation result and the deformed inner layer to obtain a fused image includes:

[0027] According to a preset arrangement order, the modules included in the segmentation result and the deformed inner layer are subjected to layer superposition processing to obtain the fused image.

[0028] Optionally, the method further includes:

[0029] Determining whether the third image meets a preset aesthetic condition;

[0030] If the third image meets the aesthetic condition, the third image is output.

[0031] Optionally, the aesthetic conditions include:

[0032] The target position of the outer garment and the target position of the inner garment are consistent; and / or,

[0033] The inner clothing was not damaged.

[0034] Optionally, the method further includes:

[0035] Determining whether the inner garment in the first image is the front view;

[0036] In the case where the inner garment is facing forward, the step of deforming the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body is performed.

[0037] According to a second aspect of the present disclosure, there is provided an image processing method, comprising:

[0038] Providing a first upload interface and a second upload interface;

[0039] In response to an operation of matching inner and outer garments, obtaining a first image uploaded by the user through the first uploading interface and including the inner garment, and a second image uploaded by the user through the second uploading interface and including a person wearing the outer garment;

[0040] performing deformation processing on the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body;

[0041] performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body;

[0042] obtaining, based on the segmentation result and the deformed inner garment, a third image of a human body wearing the inner garment and the outer garment;

[0043] The third image is displayed.

[0044] According to a third aspect of the present disclosure, an image processing method is provided, comprising:

[0045] Providing at least one innerwear image including innerwear, and at least one outerwear image including a person wearing outerwear;

[0046] In response to an operation of matching an inner garment and an outer garment, acquiring a first image selected from the inner garment images and a second image selected from the outer garment images;

[0047] performing deformation processing on the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body;

[0048] performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body;

[0049] obtaining, based on the segmentation result and the deformed inner garment, a third image of a human body wearing the inner garment and the outer garment;

[0050] The third image is displayed in the store.

[0051] According to a fourth aspect of the present disclosure, there is provided an image processing apparatus, comprising:

[0052] An image acquisition module, configured to acquire a first image including inner clothing and a second image including a person wearing outer clothing;

[0053] an inner garment deformation module, configured to deform the inner garment in the first image according to the second image, to obtain a deformed inner garment that matches the posture of the human body;

[0054] a human body segmentation module, configured to perform a first segmentation process on the human body in the second image to obtain a segmentation result of the human body;

[0055] a fusion processing module, configured to obtain, based on the segmentation result and the deformed inner garment, a third image of a human body wearing the inner garment and the outer garment;

[0056] An image output module is configured to output the third image.

[0057] According to a fifth aspect of the present disclosure, there is provided an electronic device, including:

[0058] The device according to the fourth aspect of the present disclosure; or,

[0059] A processor and a memory, wherein the memory is used to store executable instructions, and the instructions are used to control the processor to execute the method according to the first aspect to the third aspect of the present disclosure.

[0060] According to a sixth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method according to the first to third aspects of the present disclosure is implemented.

[0061] Through the image processing method of this embodiment, by processing a first image including inner clothing and a second image including a human body wearing outer clothing to obtain a third image, it is possible to achieve the technical effect of virtual fitting of a beautiful and realistic human body in the generated third image, and the human body wearing inner clothing and outer clothing can be displayed to the user through the third image.

[0062] Furthermore, the second image in this embodiment can use a real model image as the original source material. Therefore, during the processing of the first and second images, the texture of the garment can be effectively preserved without any confusion. Furthermore, the use of a real model image ensures that the present disclosure is universally applicable to various specially designed styles, preventing errors in the alignment of parts such as collars and sleeves.

[0063] Furthermore, the present invention can use a first image of an inner garment in any front view and any pose, and a second image of a person in any front view and any standing position, to generate a virtual fitting effect. This allows users to directly reuse pre-photographed model photos, eliminating the need to re-produce new material, resulting in lower costs. Furthermore, using real-life model images ensures that the virtual fitting effect achieved in this solution is as realistic as a real photo.

[0064] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0066] Figure 1 is a block diagram of an example of a hardware configuration of an electronic device that can be used to implement an embodiment of the present disclosure.

[0067] Figure 2 is a block diagram of another example of a hardware configuration of an electronic device that can be used to implement an embodiment of the present disclosure;

[0068] Figure 3is a schematic diagram of a scene of an image processing method according to an embodiment of the present disclosure;

[0069] Figure 4 is a flowchart of an image processing method according to a first embodiment of the present disclosure;

[0070] Figure 5 is a schematic diagram of an example of an image processing method according to the first embodiment of the present disclosure;

[0071] Figure 6 is a flowchart of an image processing method according to a second embodiment of the present disclosure;

[0072] Figure 7 is a flowchart of an image processing method according to a third embodiment of the present disclosure;

[0073] Figure 8 is a principle block diagram of an image processing device according to an embodiment of the present disclosure;

[0074] Figure 9 Schematic diagram of the hardware structure of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0075] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present disclosure.

[0076] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0077] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0078] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0079] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0080] <Hardware Configuration>

[0081] Figure 1 and Figure 2 This is a block diagram of a hardware configuration of an electronic device that can be used to implement the image processing method of any embodiment of the present disclosure.

[0082] In one embodiment, Figure 1 As shown, the electronic device is a server 1100.

[0083] The server 1100 provides equipment for processing, database, and communication facilities. The server 1100 can be a monolithic server or a distributed server across multiple computers or computer data centers. The server can be of various types, such as, but not limited to, a web server, a news server, a mail server, a message server, an advertising server, a file server, an application server, an interactive server, a database server, or a proxy server. In some embodiments, each server can include hardware, software, or an embedded logic component for performing the appropriate functions supported or implemented by the server, or a combination of two or more such components. For example, the server can be a blade server, a cloud server, etc., or can be a server group consisting of multiple servers, which can include one or more of the above-mentioned types of servers, etc.

[0084] In this embodiment, the server 1100 can be as follows Figure 1 As shown, it includes a processor 1110 , a memory 1120 , an interface device 1130 , a communication device 1140 , a display device 1150 , and an input device 1160 .

[0085] In this embodiment, the server 1100 may also include a speaker, a microphone, etc., which are not limited here.

[0086] The processor 1110 can be a dedicated server processor, or a desktop processor or mobile processor that meets the performance requirements, and is not limited here. The memory 1120 includes, for example, ROM (read-only memory), RAM (random access memory), and non-volatile memory such as a hard disk. The interface device 1130 includes, for example, various bus interfaces, such as a serial bus interface (including a USB interface), a parallel bus interface, etc. The communication device 1140 is capable of wired or wireless communication. The display device 1150 is, for example, a liquid crystal display, an LED display, a touch display, etc. The input device 1160 may include, for example, a touch screen, a keyboard, etc.

[0087] In this embodiment, the memory 1120 of the server 1100 is used to store instructions for controlling the processor 1110 to perform at least the image processing method according to any embodiment of the present disclosure. A skilled person can design instructions based on the disclosed solutions. How instructions control the processor's operations is well known in the art and will not be described in detail here.

[0088] Despite Figure 1, multiple devices of the server 1100 are shown; however, the present disclosure may only involve some of the devices, for example, the server 1100 only involves the memory 1120 and the processor 1110 .

[0089] In one embodiment, the electronic device 1000 may be a terminal device 1200 such as a PC or a laptop computer used by an operator, which is not limited here.

[0090] In another embodiment, Figure 2 As shown, the electronic device is a terminal device 1200.

[0091] The terminal device 1200 may include a processor 1210 , a memory 1220 , an interface device 1230 , a communication device 1240 , a display device 1250 , an input device 1260 , a speaker 1270 , a microphone 1280 , and the like.

[0092] The processor 1210 may be a mobile processor. The memory 1220 may include, for example, ROM (read-only memory), RAM (random access memory), and non-volatile memory such as a hard disk. The interface device 1230 may include, for example, a USB interface, a headphone jack, and the like. The communication device 1240 may be capable of wired or wireless communication. The communication device 1240 may include a short-range communication device, such as any device that performs short-range wireless communication based on a short-range wireless communication protocol such as Hilink, WiFi (IEEE 802.11), Mesh, Bluetooth, ZigBee, Thread, Z-Wave, NFC, UWB, LiFi, or the like. The communication device 1240 may also include a long-range communication device, such as any device that performs WLAN, GPRS, or 2G / 3G / 4G / 5G long-range communication. The display device 1250 may be, for example, an LCD display or a touch screen display. The input device 1260 may include, for example, a touch screen or a keyboard. The user may input / output voice information through the speaker 1270 and microphone 1280.

[0093] In this embodiment, the memory 1220 of the terminal device 1200 is used to store instructions for controlling the processor 1210 to perform at least the image processing method according to any embodiment of the present disclosure. A skilled person can design instructions based on the disclosed solutions. How instructions control the processor's operations is well known in the art and will not be described in detail here.

[0094] Despite Figure 2 , multiple devices of the terminal device 1200 are shown, but the present disclosure may only involve some of the devices, for example, the terminal device 1200 only involves the memory 1220, the processor 1210, and the display device 1250.

[0095] <Application Scenario>

[0096] Figure 3 Schematic diagram of an application scenario of the image processing method according to an embodiment of the present disclosure.

[0097] like Figure 3 As shown, the first image includes inner clothing, and the second image includes a human body wearing outer clothing.

[0098] In this embodiment, the electronic device processes the first and second images to obtain and output a third image. The third image includes a human body, wherein the human body is fully clothed with the inner garment in the first image and the outer garment in the second image, with the inner garment worn within the outer garment. In one example, the third image can be output by displaying it in an interface or by outputting it to another electronic device.

[0099] Specifically, the electronic device may process the first image and the second image in such a way that the electronic device processes the first image and the second image, including: deforming the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body; performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body; and obtaining a third image based on the segmentation result and the deformed inner garment.

[0100] In this way, through the image processing method of this embodiment, by processing the first image containing inner clothing and the second image containing the human body wearing outer clothing to obtain a third image, it is possible to achieve the technical effect of virtual fitting of a beautiful and real human body in the generated third image, and the human body wearing inner clothing and outer clothing can be displayed to the user through the third image.

[0101] Furthermore, the second image in this embodiment can use a real model image as the original source material. Therefore, during the processing of the first and second images, the texture of the garment can be effectively preserved without any confusion. Furthermore, the use of a real model image ensures that the present disclosure is universally applicable to various specially designed styles, preventing errors in the alignment of parts such as collars and sleeves.

[0102] Furthermore, the present invention can use a first image of an inner garment in any front view and any pose, and a second image of a person in any front view and any standing position, to generate a virtual fitting effect. This allows users to directly reuse pre-photographed model photos, eliminating the need to re-produce new material, resulting in lower costs. Furthermore, using real-life model images ensures that the virtual fitting effect achieved in this solution is as realistic as a real photo.

[0103] <Method Example>

[0104] Figure 4 is a flow chart of an image processing method according to an embodiment of the present disclosure, which can be implemented by an electronic device. The electronic device can be Figure 1 The server 1100 shown or Figure 2 The terminal device 1200 is shown.

[0105] according to Figure 4 As shown, the image processing method of this embodiment may include the following steps S4100 to S4500:

[0106] Step S4100: Acquire a first image including inner clothing and a second image including a human body wearing outer clothing.

[0107] In this embodiment, inner clothing refers to clothing that can be worn inside outer clothing relative to outer clothing.

[0108] The first image and the second image in this embodiment can be images selected by the user from the network, or downloaded by the user from the network to the electronic device that executes the method of this embodiment, or uploaded by the user from other electronic devices to the electronic device that executes the method of this embodiment.

[0109] In one example, the electronic device may pre-display multiple images containing inner garments and multiple images containing a person wearing outer garments. The electronic device may also provide a matching button, allowing the user to select one of the multiple images containing inner garments as a first image and one of the multiple images containing a person wearing outer garments as a second image based on their preference. By clicking the matching button, the electronic device is triggered to execute the method of this embodiment.

[0110] In this embodiment, the second image can be a real model picture, and the human body can be a picture of the model wearing outerwear. Specifically, the human body can be a full-body human body or a half-body human body, which is not limited here.

[0111] Step S4200: deform the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body.

[0112] In this embodiment, the inner garment is deformed so that the deformed inner garment has the same posture as the human body in the second image, so that the inner garment can be worn on the human body.

[0113] In one embodiment of the present disclosure, before executing step S4200, the method may further include: determining whether the inner garment in the first image is facing forward; if the inner garment in the first image is facing forward, it indicates that the inner garment can be normally worn inside the outer garment worn by the human body, and therefore, step S4200 may be executed; if the inner garment in the first image is not facing forward, it indicates that the inner garment cannot be normally worn inside the outer garment worn by the human body, and therefore, execution of steps S4200 to S4500 of this embodiment is stopped, and a prompt message indicating that the inner garment in the first image is not facing forward may be output.

[0114] Specifically, whether the inner garment is front-facing may be determined based on the collar and / or front placket of the inner garment.

[0115] In this embodiment, a first neural network model may be used to determine whether the inner garment in the first image is facing the front. Specifically, the first image may be input into the first neural network model, and a determination result of whether the inner garment in the first image is facing the front may be output.

[0116] In one embodiment of the present disclosure, deformation processing is performed on the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body, which may include the following steps S4210 to S4240:

[0117] Step S4210: performing a second segmentation process on the inner garment in the first image to obtain a front piece of the inner garment.

[0118] In this embodiment, the inner garment in the first image may be first extracted, and then the second segmentation process may be performed on the inner garment, and the segmentation result may include the neckline, the front piece, and the back hem. In this embodiment, only the front piece of the inner garment may be obtained.

[0119] In one example, a second segmentation process may be performed on the inner garment in the first image based on a preset second neural network model to obtain the front piece of the inner garment. Specifically, the first image may be input into the second neural network model to obtain the front piece of the inner garment in the first image.

[0120] Step S4220: extracting the key points of the inner garment in the first image.

[0121] The inner wear key points in this embodiment can be used to represent the outline of the inner wear clothing. For example, the inner wear key points can include the shoulders, neckline, cuffs, hem, front and other positions of the inner wear clothing.

[0122] In one example, the key points of the inner garment in the first image may be extracted based on a preset third neural network model. Specifically, the first image may be input into the third neural network model to obtain the key points of the inner garment in the first image.

[0123] Step S4230: extracting key points of the human body in the second image.

[0124] The key points of the human body in this embodiment can be used to represent the outline of the human body. For example, the key points of the inner wear can include the shoulders, elbows, armpits, waist and other positions of the human body.

[0125] In one example, the human body key points of the human body in the second image may be extracted based on a preset fourth neural network model. Specifically, the second image may be input into the fourth neural network model to obtain the human body key points of the human body in the second image.

[0126] Step S4240: Obtain a deformed inner garment based on the front piece of the inner garment, the human body in the second image, the inner garment key points, and the human body key points.

[0127] In one embodiment of the present disclosure, obtaining a deformed inner garment based on the front piece of the inner garment, the human body in the second image, the inner garment key points, and the human body key points may include the following steps S4241 to S4243:

[0128] Step S4241: obtaining a first projection matrix based on the front piece of the inner garment and a preset reference human body, and obtaining a second projection matrix based on the human body in the second image and the reference human body.

[0129] In this embodiment, the first projection matrix may be a projection matrix that enables the front piece of the inner garment to fit the reference human body, and the second projection matrix may be a projection matrix that enables the human body in the second image to fit the reference human body. Fit in this embodiment means that the posture is the same.

[0130] In one example, a first projection matrix can be obtained based on a preset fifth neural network model according to the front piece of the inner garment and a reference human body; and a second projection matrix can be obtained based on a preset sixth neural network model according to the human body in the second image and the reference human body.

[0131] Specifically, the front piece of the inner garment and the reference human body can be input into the fifth neural network model to obtain the first projection matrix; the human body in the second image (or the second image) and the reference human body can be input into the sixth neural network model to obtain the second projection matrix.

[0132] Step S4242: Process the front piece of the inner garment according to the first projection matrix to obtain the inner garment projection; and process the human body in the second image according to the second projection matrix to obtain the human body projection.

[0133] In this embodiment, applying the first projection matrix to the front piece of the inner garment can obtain the projection of the front piece of the inner garment on the reference body, i.e., the inner garment projection. Applying the second projection matrix to the body in the second image can obtain the projection of the body in the second image on the reference body, i.e., the body projection.

[0134] Step S4243: Obtain a deformed inner garment according to the inner garment projection, the human body projection, the inner garment key points, and the human body key points.

[0135] In this embodiment, the inner garment projection may represent the positions of the key points of the inner garment on the reference body, and the body projection may represent the positions of the key points of the human body on the reference body. The inner garment projection is deformed based on the inner garment projection, the human body projection, the key points of the inner garment, and the key points of the human body to obtain a deformed inner garment.

[0136] In one example, a deformed inner garment can be obtained based on the inner garment projection, the human body projection, the inner garment key points, and the human body key points based on a preset seventh neural network model. Specifically, the inner garment projection, the human body projection, the inner garment key points, and the human body key points can be input into the seventh neural network model to obtain the deformed inner garment.

[0137] Step S4300: performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body.

[0138] In this embodiment, the human body in the second image may be extracted first, and then the first segmentation process may be performed on the human body. The obtained segmentation result may include hair, face, neck + chest, inner wear area, coat + other human body areas.

[0139] In one example, a first segmentation process may be performed on the human body in the second image based on the preset eighth neural network model to obtain the front piece of the inner garment. Specifically, the second image may be input into the eighth neural network model to obtain a segmentation result of the human body in the second image.

[0140] Step S4400: Obtain a third image of the human body including the inner garment and the outer garment based on the segmentation result and the deformed inner garment.

[0141] In one embodiment of the present disclosure, obtaining a third image of a person wearing inner clothing and outer clothing based on the segmentation result and the deformed inner clothing may include steps S4410 to S4420 as follows:

[0142] Step S4410: Fusing the segmentation result and the deformed image to obtain a fused image.

[0143] In this embodiment, the inner layer area in the segmentation result can be replaced first according to the deformed inner layer, and then the position of each module in the segmentation result in the human body can be calculated, and the modules included in the segmentation result can be layered according to a preset arrangement order to obtain a fused image.

[0144] Specifically, in the order of layers from bottom to top, they can be neck + chest, face, deformed inner wear, coat + other body areas, and hair.

[0145] In one embodiment of the present disclosure, because the neckline height of the deformed inner garment differs from the neckline height of the inner garment region on the human body, directly replacing the inner garment region in the segmentation result based on the deformed inner garment may cause damage to the human body in the third image. Therefore, the method may further include: performing skin repair processing on the inner garment region in the segmentation result.

[0146] In this embodiment, the skin repair treatment is performed on the inner wear area, which may be to adjust the collar height of the inner wear area to a specified height, that is, to repair all parts of the inner wear area above the specified height to skin.

[0147] The repaired segmentation result can include the neck + chest, face, coat + other body parts, hair, and the repaired inner wear area. Among them, the neck + chest, face, coat + other body parts, and hair modules in the repaired segmentation result are unchanged from the original segmentation result.

[0148] Step S4420: Erasing the portion of the deformed inner garment that exceeds the human body in the fused image to obtain a third image.

[0149] Since it is necessary to reflect the effect of wearing inner clothing on the human body, and the posture of the inner clothing may be different from that of the human body, it is necessary to erase the part of the deformed inner clothing that exceeds the body range of the human body in the fused image. In this way, the wearing effect of the inner clothing in the third image can be made more realistic.

[0150] Step S4500: output the third image.

[0151] In this embodiment, the method of outputting the third image may include: transmitting the third image to other electronic devices, and / or displaying the third image in an interface.

[0152] In this way, through the image processing method of this embodiment, by processing the first image containing inner clothing and the second image containing the human body wearing outer clothing to obtain a third image, it is possible to achieve the technical effect of virtual fitting of a beautiful and real human body in the generated third image, and the human body wearing inner clothing and outer clothing can be matched, and the user can be shown the matching effect of the inner clothing and outer clothing through the third image.

[0153] Furthermore, the second image in this embodiment can use a real model image as the original source material. Therefore, during the processing of the first and second images, the texture of the garment can be effectively preserved without any confusion. Furthermore, the use of a real model image ensures that the present disclosure is universally applicable to various specially designed styles, preventing errors in the alignment of parts such as collars and sleeves.

[0154] Furthermore, the present invention can use a first image of an inner garment in any front view and any pose, and a second image of a person in any front view and any standing position, to generate a virtual fitting effect. This allows users to directly reuse pre-photographed model photos, eliminating the need to re-produce new material, resulting in lower costs. Furthermore, using real-life model images ensures that the virtual fitting effect achieved in this solution is as realistic as a real photo.

[0155] In one embodiment of the present disclosure, based on obtaining the third image, the method may further include steps S4600 to S4700 as follows:

[0156] Step S4600: Determine whether the third image meets the preset aesthetic conditions.

[0157] In this embodiment, the aesthetic condition may be set in advance according to the application scenario or specific needs.

[0158] In one example, aesthetic conditions may include:

[0159] The target position of the outer garment matches the target position of the inner garment; and / or,

[0160] The inner clothing was not damaged.

[0161] In this embodiment, whether the third image meets the aesthetic condition may be determined based on the preset ninth neural network model, and a determination result of whether the third image meets the aesthetic condition may be output.

[0162] Step S4700: If it is determined that the third image meets the aesthetic condition, output the third image.

[0163] If it is determined that the third image does not meet the aesthetic condition, the third image may not be output. Further, a prompt message indicating that the third image does not meet the aesthetic condition may be output.

[0164] In this way, it can be ensured that the output image meets the aesthetic conditions, and the matching effect of the inner garment and the outer garment in the output third image is improved.

[0165] <Example>

[0166] Figure 5 is a schematic diagram of an example of an image processing method according to an embodiment of the present disclosure.

[0167] like Figure 5 As shown, the method may include: acquiring a first image including inner clothing; extracting inner clothing key points of the inner clothing in the first image; performing a second segmentation process on the inner clothing in the first image to obtain a front piece of the inner clothing; obtaining a first projection matrix based on the front piece of the inner clothing and a preset reference human body; processing the front piece of the inner clothing according to the first projection matrix to obtain an inner clothing projection; acquiring a second image including a human body wearing outer clothing; extracting human body key points of the human body in the second image; obtaining a second projection matrix based on the human body in the second image and the reference human body; processing the human body in the second image according to the second projection matrix to obtain a human body projection; obtaining a deformed inner clothing based on the inner clothing projection, the human body projection, the inner clothing key points, and the human body key points; performing a first segmentation process on the human body in the second image to obtain a human body segmentation result; performing skin repair processing on the inner clothing area of the segmentation result; performing a fusion process on the segmentation result and the deformed inner clothing to obtain a fused image; erasing the portion of the deformed inner clothing in the fused image that exceeds the human body range to obtain a third image.

[0168] <Second Method Example>

[0169] Figure 6 is a flow chart of an image processing method according to a second embodiment of the present disclosure, which can be implemented by an electronic device. The electronic device can be as follows: Figure 1 The server 1100 shown or Figure 2 The terminal device 1200 is shown.

[0170] according to Figure 6 As shown, the image processing method of this embodiment may include the following steps S6100 to S6600:

[0171] Step S6100: Provide a first upload interface and a second upload interface.

[0172] In this embodiment, the user can click on the first upload interface to trigger the electronic device to display multiple innerwear images of people wearing innerwear and a first upload button. The user can select an innerwear image according to their own preferences and click the first upload button to upload the innerwear image selected by the user. The user can click on the second upload interface to trigger the electronic device to display multiple outerwear images of people wearing outerwear and a second upload button. The user can select an outerwear image according to their own preferences and click the second upload button to upload the outerwear image selected by the user.

[0173] Step S6200, in response to the operation of matching inner garments and outer garments, obtain a first image containing the inner garments uploaded by the user through the first upload interface, and a second image containing the human body wearing the outer garments uploaded by the user through the second upload interface.

[0174] In this embodiment, a matching button may be provided, and the operation of the user clicking the matching button may be used as an operation of matching inner clothing and outer clothing.

[0175] Furthermore, the inner wear image uploaded by the user through the first interface may be used as the first image, and the outer wear image uploaded by the user through the second interface may be used as the second image.

[0176] Step S6300: deform the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body.

[0177] The step S6300 can refer to the step S4200 in the first method embodiment, and will not be described in detail here.

[0178] Step S6400: performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body.

[0179] The step S6400 can refer to the step S4300 in the first method embodiment, and will not be described in detail here.

[0180] Step S6500: Obtain a third image of the human body wearing the inner garment and the outer garment according to the segmentation result and the deformed inner garment.

[0181] The step S6500 can refer to the step S4400 in the first method embodiment, and will not be described in detail here.

[0182] Step S6600: display the third image.

[0183] The method of this embodiment processes a first image containing inner clothing and a second image containing a person wearing outer clothing to generate a third image. This can achieve the technical effect of a virtual fitting of a person wearing both inner and outer clothing in a beautiful and realistic third image. Furthermore, the third image can be used to show the user the effect of the inner and outer clothing being combined. The third image can be used to provide a user with a reference for the matching effect of the inner and outer clothing.

[0184] Furthermore, the second image in this embodiment can use a real model image as the original source material. Therefore, during the processing of the first and second images, the texture of the garment can be effectively preserved without any confusion. Furthermore, the use of a real model image ensures that the present disclosure is universally applicable to various specially designed styles, preventing errors in the alignment of parts such as collars and sleeves.

[0185] Furthermore, the present invention can use a first image of an inner garment in any front view and any pose, and a second image of a person in any front view and any standing position, to generate a virtual fitting effect. This allows users to directly reuse pre-photographed model photos, eliminating the need to re-produce new material, resulting in lower costs. Furthermore, using real-life model images ensures that the virtual fitting effect achieved in this solution is as realistic as a real photo.

[0186] <Third Method Embodiment>

[0187] Figure 7 is a flow chart of an image processing method according to a third embodiment of the present disclosure, which can be implemented by an electronic device. The electronic device can be Figure 1 The server 1100 shown or Figure 2 The terminal device 1200 is shown.

[0188] according to Figure 7 As shown, the image processing method of this embodiment may include the following steps S7100 to S7600:

[0189] Step S7100: providing at least one innerwear image including innerwear, and at least one outerwear image including a person wearing outerwear in the store.

[0190] In this embodiment, the store may be a virtual seller's store, and images of the products sold in the store may be displayed in the store.

[0191] Step S7200: In response to the operation of matching the inner garment and the outer garment, a first image selected from the inner garment images and a second image selected from the outer garment images are acquired.

[0192] In this embodiment, in the interface for providing inner and outer clothing in the store, a selection box may be provided on each inner and outer clothing, and the user may select any inner clothing and any outer clothing according to his or her own preferences, and perform the operation of matching the inner and outer clothing by clicking the matching button, triggering the electronic device to obtain the inner image selected by the user as the first image and the outer image selected by the user as the second image.

[0193] Step S7300: deform the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body.

[0194] The step S7300 can refer to the step S4200 in the first method embodiment, and will not be described in detail here.

[0195] Step S7400: performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body.

[0196] The step S7400 can refer to the step S4300 in the first method embodiment, and will not be described in detail here.

[0197] Step S7500: obtaining a third image of a human body wearing inner clothing and outer clothing according to the segmentation result and the deformed inner clothing.

[0198] The step S7500 can refer to the step S4400 in the first method embodiment, and will not be described in detail here.

[0199] Step S7600: display the third image in the store.

[0200] The method of this embodiment processes a first image containing inner clothing and a second image containing a person wearing outer clothing to generate a third image. This generates a third image that contains a beautiful, realistic person wearing both inner and outer clothing, creating a virtual fitting effect. The third image then demonstrates to the user the effect of the inner and outer clothing being combined. Displaying the third image in a store allows users to gain a reference for the matching effect of the inner and outer clothing.

[0201] Furthermore, the second image in this embodiment can use a real model image as the original source material. Therefore, during the processing of the first and second images, the texture of the garment can be effectively preserved without any confusion. Furthermore, the use of a real model image ensures that the present disclosure is universally applicable to various specially designed styles, preventing errors in the alignment of parts such as collars and sleeves.

[0202] Furthermore, the present invention can use a first image of an inner garment in any front view and any pose, and a second image of a person in any front view and any standing position, to generate a virtual fitting effect. This allows users to directly reuse pre-photographed model photos, eliminating the need to re-produce new material, resulting in lower costs. Furthermore, using real-life model images ensures that the virtual fitting effect achieved in this solution is as realistic as a real photo.

[0203] <Device Example>

[0204] Figure 8 FIG. 8 is a principle block diagram of an image processing apparatus 8000 according to an embodiment of the present disclosure. Figure 8 As shown, the image processing device 8000 may include an image acquisition module 8100, an inner wear deformation module 8200, a human body segmentation module 8300, a fusion processing module 8400, and an image output module 8500. The image acquisition module 8100 is used to acquire a first image containing inner clothing and a second image containing a human body wearing outer clothing; the inner wear deformation module 8200 is used to deform the inner clothing in the first image according to the second image to obtain a deformed inner clothing that matches the posture of the human body; the human body segmentation module 8300 is used to perform a first segmentation process on the human body in the second image to obtain a segmentation result of the human body; the fusion processing module 8400 is used to obtain a third image containing a human body wearing the inner clothing and the outer clothing according to the segmentation result and the deformed inner clothing; and the image output module 8500 is used to output the third image.

[0205] In one embodiment of the present disclosure, the inner deformation module 8200 is further configured to:

[0206] performing a second segmentation process on the inner garment in the first image to obtain a front piece of the inner garment;

[0207] extracting key points of the inner garment in the first image;

[0208] Extracting key points of the human body in the second image;

[0209] The deformed inner garment is obtained according to the front piece of the inner garment, the human body in the second image, the key points of the inner garment and the key points of the human body.

[0210] In one embodiment of the present disclosure, obtaining the deformed inner garment according to the front piece of the inner garment, the human body in the second image, the inner garment key points, and the human body key points includes:

[0211] Obtaining a first projection matrix based on the front piece of the inner garment and a preset reference human body, and obtaining a second projection matrix based on the human body in the second image and the reference human body;

[0212] Processing the front piece of the inner garment according to the first projection matrix to obtain an inner garment projection; processing the human body according to the second projection matrix to obtain a human body projection;

[0213] The deformed inner layer is obtained according to the inner layer projection and the human body projection, the inner layer key points and the human body key points.

[0214] In one embodiment of the present disclosure, the fusion processing module 8400 is further configured to:

[0215] Fusing the segmentation result and the deformed inner layer to obtain a fused image;

[0216] The portion of the deformed inner garment that exceeds the body range of the human body in the fused image is erased to obtain the third image.

[0217] In one embodiment of the present disclosure, the image processing apparatus 8000 further includes:

[0218] The skin repair module is used to perform skin repair processing on the inner area in the segmentation result.

[0219] In one embodiment of the present disclosure, fusing the segmentation result and the deformed inner layer to obtain a fused image includes:

[0220] According to a preset arrangement order, the modules included in the segmentation result and the deformed inner layer are subjected to layer superposition processing to obtain the fused image.

[0221] In one embodiment of the present disclosure, the image processing apparatus 8000 further includes:

[0222] an aesthetic judgment module, configured to judge whether the third image meets a preset aesthetic condition;

[0223] The image output module 8500 is further configured to output the third image if the third image meets the aesthetic condition.

[0224] In one embodiment of the present disclosure, the aesthetic conditions include:

[0225] The target position of the outer garment and the target position of the inner garment are consistent; and / or,

[0226] The inner clothing was not damaged.

[0227] In one embodiment of the present disclosure, the image processing apparatus 8000 further includes:

[0228] a front-facing judgment module, configured to judge whether the inner garment in the first image is facing the front;

[0229] The inner wear deformation module 8200 is used to execute the step of deforming the inner wear in the first image according to the second image to obtain a deformed inner wear that matches the posture of the human body when the inner wear is facing forward.

[0230] Those skilled in the art will appreciate that the image processing apparatus 8000 can be implemented in various ways. For example, the image processing apparatus 8000 can be implemented by configuring a processor with instructions. For example, instructions can be stored in a ROM, and when the device is started, the instructions are read from the ROM into a programmable device to implement the image processing apparatus 8000. For example, the image processing apparatus 8000 can be embedded in a dedicated device (e.g., an ASIC). The image processing apparatus 8000 can be divided into independent units, or they can be combined together for implementation. The image processing apparatus 8000 can be implemented by one of the various implementations described above, or by a combination of two or more of the various implementations described above.

[0231] In this embodiment, the image processing device 8000 can have multiple implementation forms. For example, the image processing device 8000 can be a functional module running in any software product or application that provides image processing services, or it can be a peripheral embedded component, plug-in, patch, etc. of these software products or applications, or it can be these software products or applications themselves.

[0232] <Electronic equipment>

[0233] In this embodiment, an electronic device 1000 is further provided. The electronic device 1000 may be Figure 1 The server 1100 shown may also be Figure 2 The terminal device 1200 is shown.

[0234] In one aspect, the electronic device 1000 may include the image processing apparatus 8000 according to any embodiment of the present disclosure, for implementing the image processing method according to any embodiment of the present disclosure.

[0235] On the other hand, Figure 9 As shown, the electronic device 1000 may further include a processor 1300 and a memory 1400, wherein the memory 1400 is used to store executable instructions; the processor 1300 is used to control the operation of the electronic device 1000 to execute the image processing method according to any embodiment of the present disclosure according to the instructions.

[0236] <Computer-readable storage medium>

[0237] In this embodiment, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the image processing method of any embodiment of the present disclosure is implemented.

[0238] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0239] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0240] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0241] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages such as Smalltalk, C++, and conventional procedural programming languages such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0242] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0243] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0244] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0245] The flowcharts and block diagrams in the accompanying drawings show the possible implementation architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of an instruction, and the module, program segment or part of the instruction contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are all equivalent.

[0246] The embodiments of the present disclosure have been described above. The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, their practical applications, or technical improvements in the marketplace, or to enable other persons skilled in the art to understand the embodiments disclosed herein. The scope of the present disclosure is defined by the appended claims.

Claims

1. An image processing method, wherein: include: Acquire a first image including inner clothing and a second image including a person wearing outer clothing; performing deformation processing on the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body; performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body, wherein the segmentation result includes an inner layer area; obtaining, based on the segmentation result and the deformed inner garment, a third image of a human body wearing the inner garment and the outer garment; outputting the third image, Among them, obtaining a third image of a human body wearing the inner garment and the outer garment based on the segmentation result and the deformed inner garment includes: replacing the inner garment area in the segmentation result according to the deformed inner garment.

2. The method according to claim 1, wherein deforming the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body comprises: performing a second segmentation process on the inner garment in the first image to obtain a front piece of the inner garment; extracting key points of the inner garment in the first image; Extracting key points of the human body in the second image; The deformed inner garment is obtained according to the front piece of the inner garment, the human body in the second image, the key points of the inner garment and the key points of the human body.

3. The method according to claim 2, wherein obtaining the deformed inner garment based on the front piece of the inner garment, the human body in the second image, the inner garment key points, and the human body key points comprises: Obtaining a first projection matrix based on the front piece of the inner garment and a preset reference human body, and obtaining a second projection matrix based on the human body in the second image and the reference human body; Processing the front piece of the inner garment according to the first projection matrix to obtain an inner garment projection; processing the human body according to the second projection matrix to obtain a human body projection; The deformed inner layer is obtained according to the inner layer projection and the human body projection, the inner layer key points and the human body key points.

4. The method according to claim 1, wherein obtaining a third image of a person wearing the inner garment and the outer garment based on the segmentation result and the deformed inner garment comprises: Fusing the segmentation result and the deformed inner layer to obtain a fused image; The portion of the deformed inner garment that exceeds the body range of the human body in the fused image is erased to obtain the third image.

5. The method according to claim 4, further comprising: Perform skin repair processing on the inner layer area in the segmentation result.

6. The method according to claim 4, wherein fusing the segmentation result and the deformed inner layer to obtain a fused image comprises: According to a preset arrangement order, the modules included in the segmentation result and the deformed inner layer are subjected to layer superposition processing to obtain the fused image.

7. The method according to claim 5, further comprising: Determining whether the third image meets a preset aesthetic condition; If the third image meets the aesthetic condition, the third image is output.

8. The method according to claim 7, wherein the aesthetic condition comprises: The target position of the outer garment matches the target position of the inner garment; and / or, The inner clothing was not damaged.

9. The method according to claim 1, further comprising: Determining whether the inner garment in the first image is the front view; In the case where the inner garment is facing forward, the step of deforming the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body is performed.

10. An image processing method, wherein: include: Providing a first upload interface and a second upload interface; In response to an operation of matching inner and outer garments, obtaining a first image uploaded by the user through the first uploading interface and including the inner garment, and a second image uploaded by the user through the second uploading interface and including a person wearing the outer garment; performing deformation processing on the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body; performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body, wherein the segmentation result includes an inner layer area; obtaining, based on the segmentation result and the deformed inner garment, a third image of a human body wearing the inner garment and the outer garment; showing the third image, Among them, obtaining a third image of a human body wearing the inner garment and the outer garment based on the segmentation result and the deformed inner garment includes: replacing the inner garment area in the segmentation result according to the deformed inner garment.

11. An image processing method, wherein: include: Providing at least one innerwear image including innerwear, and at least one outerwear image including a person wearing outerwear; In response to an operation of matching an inner garment and an outer garment, acquiring a first image selected from the inner garment images and a second image selected from the outer garment images; performing deformation processing on the inner garment in the first image according to the second image to obtain a deformed inner garment that matches the posture of the human body; performing a first segmentation process on the human body in the second image to obtain a segmentation result of the human body, wherein the segmentation result includes an inner layer area; obtaining, based on the segmentation result and the deformed inner garment, a third image of a human body wearing the inner garment and the outer garment; displaying the third image in the store, Among them, obtaining a third image of a human body wearing the inner garment and the outer garment based on the segmentation result and the deformed inner garment includes: replacing the inner garment area in the segmentation result according to the deformed inner garment.

12. An image processing device, comprising: An image acquisition module, configured to acquire a first image including inner clothing and a second image including a person wearing outer clothing; an inner garment deformation module, configured to deform the inner garment in the first image according to the second image, to obtain a deformed inner garment that matches the posture of the human body; a human body segmentation module, configured to perform a first segmentation process on the human body in the second image to obtain a segmentation result of the human body, wherein the segmentation result includes an inner layer area; a fusion processing module, configured to obtain, based on the segmentation result and the deformed inner garment, a third image of a human body wearing the inner garment and the outer garment; an image output module, configured to output the third image, Among them, the fusion processing module is specifically used to replace the inner layer area in the segmentation result according to the deformed inner layer.

13. An electronic device, wherein: include: The device according to claim 12; or, A processor and a memory, wherein the memory is used to store executable instructions, and the instructions are used to control the processor to execute the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method according to any one of claims 1 to 11 when executed by a processor.

Images