Digital human reloading method and device, storage medium and program product

Abstract

The invention discloses a digital human body reloading method and device, a storage medium and a program product, and relates to the technical field of data processing, because a preset digital human body is created based on a dress-up model and a human body structure model which are separated from each other, a user only needs to obtain a reloading demand of the user for the preset digital human body; determining a target dress-up model meeting the dress-up requirement from a preset dress-up database; integrating the target dress-up model and the human body structure model of the preset digital person to obtain an integrated model, and creating a target digital person after dress changing based on the integrated model; according to the technical scheme, it can be understood that in the dress-up process, the human body structure model is not changed, and only the target dress-up model meeting the dress-up requirement needs to be obtained from the preset dress-up database, so that the digital human dress-up cost can be reduced while the personalized requirement of a user is met.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to digital human costume changing methods, devices, storage media and program products. Background Technology

[0002] Currently, in order to meet users' personalized digital human dressing needs, it is usually necessary to photograph a large number of dressed-up real models during the digital human creation process, and then to model based on the photographed content in order to generate a limited number of digital humans with fixed outfits.

[0003] However, since the number of digital humans with fixed outfits is limited, when the limited number of digital humans with fixed outfits cannot meet the user's dressing needs, it is necessary to reshoot real people, which increases costs. Summary of the Invention

[0004] The main purpose of this application is to provide a digital human costume-changing method, device, storage medium, and program product, aiming to solve the technical problem of how to reduce the cost of digital human costume-changing while meeting users' personalized needs.

[0005] To achieve the above objectives, this application proposes a digital human costume-changing method, the method comprising:

[0006] The system obtains the user's clothing requirements for a preset digital human, wherein the preset digital human is created based on a separate clothing model and a human body structure model.

[0007] Determine the target outfit model that meets the dressing requirements from the preset outfit database;

[0008] The target costume model and the preset digital human body structure model are integrated to obtain the integrated model;

[0009] Based on the integrated model, the target digital human after the makeover is created.

[0010] In one embodiment, the step of integrating the target costume model and the preset digital human body structure model to obtain the integrated model includes:

[0011] Based on a preset layering relationship library, it is determined whether the target costume model overlaps with the original costume model of the preset digital human. The preset layering relationship library stores the layering relationships between different costume models.

[0012] If it exists, then based on the layering relationship between the target outfit model and the original outfit model, and the preset original colors of the target outfit model and the original outfit model at the overlapping area, the target color at the overlapping area after the outfit change is determined;

[0013] Based on the target costume model, the original costume model of the preset digital human, and the target color, a new costume model is determined;

[0014] The new costume model and the preset digital human body structure model are integrated to obtain the integrated model.

[0015] In one embodiment, the step of creating the target digital human after costume change based on the integrated model includes:

[0016] Based on the target outfit type corresponding to the new outfit model and the preset outfit movement trajectory corresponding to different outfit types, the target outfit movement trajectory of the new outfit model is determined;

[0017] Based on the integrated model and the target costume's movement trajectory, a target digital human after costume change is created.

[0018] In one embodiment, before the step of obtaining the user's clothing requirements for the preset digital human, the method further includes:

[0019] Obtain raw footage of real people, including videos and / or multiple consecutive images;

[0020] Historical costumes, separated from the human anatomy of real people, were extracted from the original materials.

[0021] The historical costume is modeled to obtain the historical costume model;

[0022] Based on the historical costume model and the preset human body structure model, a preset digital human is created.

[0023] In one embodiment, after the step of extracting the historical costume separated from the human body structure of a real person from the original material, the method further includes:

[0024] Identify the pixel colors of the historical outfits;

[0025] Determine the number of pixels corresponding to different pixel colors in the historical outfit;

[0026] Based on the number of pixels and a preset threshold, the historical outfit is color-coded to obtain the original color of the historical outfit.

[0027] In one embodiment, after the step of extracting the historical costume separated from the human body structure of a real person from the original material, the method further includes:

[0028] Based on the pixel coordinates of the regions where different historical costumes are located, determine whether there are overlapping areas between the historical costumes;

[0029] Identify the true color of overlapping areas;

[0030] Based on the actual color and the original colors of at least two outfits that have overlapping areas, determine the layering relationship between the at least two outfits in the overlapping areas.

[0031] In one embodiment, after the step of extracting the historical costume separated from the human body structure of a real person from the original material, the method further includes:

[0032] Obtain the first position information of the pixel corresponding to the historical outfit in the video and / or multiple consecutive images, and determine the target outfit type that is moving based on the first position information;

[0033] Based on the second position information of the pixels corresponding to the target outfit type in the video and / or multiple consecutive images, a motion curve is fitted to obtain the motion curve.

[0034] The motion curve is used as the preset motion trajectory of the target outfit type.

[0035] In addition, to achieve the above objectives, this application also proposes a digital human costume-changing device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the digital human costume-changing method as described above.

[0036] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the digital human costume-changing method described above.

[0037] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the digital human costume-changing method described above.

[0038] One or more technical solutions proposed in this application have at least the following technical effects:

[0039] Since the preset digital human is created based on separate dressing models and human body structure models, this application only needs to obtain the user's dressing requirements for the preset digital human, determine the target dressing model that meets the dressing requirements from the preset dressing database, and integrate the target dressing model and the human body structure model of the preset digital human to obtain the integrated model. Based on the integrated model, the dressed target digital human can be created. It can be understood that the human body structure model remains unchanged during the dressing process. Only the target dressing model that meets the dressing requirements needs to be obtained from the preset dressing database. This solves the problem in related technologies that require re-photographing the dressed real model and then modeling the content as a whole, which increases costs. Therefore, this application can reduce the cost of digital human dressing while meeting the user's personalized needs. Attached Figure Description

[0040] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0041] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart illustrating an embodiment of the digital human costume-changing method of this application.

[0043] Figure 2 This is a flowchart illustrating Embodiment 2 of the digital human costume changing method of this application;

[0044] Figure 3 This is a schematic diagram of the first scene provided for Embodiment 2 of the digital human clothing change method of this application;

[0045] Figure 4 This is a schematic diagram of the second scene provided in Embodiment 2 of the digital human clothing change method of this application;

[0046] Figure 5 This is a flowchart illustrating Embodiment 3 of the digital human costume changing method of this application;

[0047] Figure 6 This is a schematic diagram of the device structure of the hardware operating environment involved in the digital human costume changing method in the embodiments of this application.

[0048] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0049] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0050] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0051] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device or digital human costume-changing device capable of performing the above functions. The following description uses a digital human costume-changing device as an example to illustrate this embodiment and the subsequent embodiments.

[0052] Based on this, the embodiments of this application provide a digital human clothing change method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the digital human costume changing method of this application.

[0053] In this embodiment, the digital human costume changing method includes steps S10 to S40:

[0054] Step S10: Obtain the user's clothing requirements for the preset digital human, wherein the preset digital human is created based on a clothing model and a human body structure model that are separate from each other;

[0055] It should be noted that digital humans are currently mainly used in interactive scenarios such as intelligent customer service and live streaming, but they can also be used in static, non-interactive scenarios where fixed text is read. Taking interactive scenarios as an example, when users interact with digital humans, they are interacting with fixed character images, and personalized configuration cannot be achieved.

[0056] To enable users to configure digital humans, it is usually necessary to photograph a large number of dressed-up real models during the creation process, and then model them based on the photographed content to generate a limited number of digital humans with fixed outfits. However, since the number of digital human outfits is limited, when the limited number of digital humans with fixed outfits cannot meet the user's dressing-up needs, it is necessary to re-photograph real people, which increases costs.

[0057] For example, to meet users' personalized digital human dressing needs, suppose we need to set 20 configurable outfits for each digital human. This would typically require 20 real-life models wearing each of the 20 outfits, and going through processes such as re-photographing and modeling real-life figures.

[0058] First, live models are filmed in a predetermined outfit, capturing their expressions, body movements, and lip movements to obtain footage. Then, based on this footage, human modeling (including outfit), expression modeling, body movement modeling, lip movement modeling, and AI (Artificial Intelligence) training are performed. Next, a digital human is created on a digital human configuration platform, configured with a preset unique ID, including the digital human's business, corporate entity, expressions, body movements, and lip movements. Finally, based on the configured or real-time generated text, corresponding audio is generated, and corresponding expressions, body movements, and lip movements of the digital human (including outfit) are generated, creating dynamic business content for the digital human to be displayed. This content is then combined into an audio-visual stream, which is pushed to the digital human interaction terminal so that users can see the digital human and hear its voice announcements, enabling further text or voice interaction with the digital human.

[0059] Understandably, in current digital human interaction scenarios, digital humans exist in the form of audio and video streams, making it impossible to achieve digital human dressing up using a layered image model. Digital humans existing in the form of audio and video streams have high requirements for later system operation. Operators need to constantly pay attention to users' demands for digital human dressing up. To meet these demands, a large number of dressed-up real-life models are photographed, and then modeling is performed based on the photographed content to generate a limited number of digital humans with fixed outfits for users to use. The above methods are time-consuming, labor-intensive, and costly.

[0060] To address the aforementioned issues, this embodiment aims to meet users' personalized needs while reducing the cost of digital human costume changes.

[0061] Specifically, the preset digital human in this embodiment is created based on a separate costume model and a human body structure model. It can be understood that the preset digital human in this embodiment is not directly modeled from a costumed real-life model, but rather the costume and human body structure model of the costumed real-life model are separated, modeled separately, and then integrated.

[0062] It is understandable that, since the preset digital avatar is created based on separate dressing models and human body structure models, users can interact with the preset digital avatar based on the currently displayed preset digital avatar image, thereby changing the preset digital avatar's outfit.

[0063] Specifically, this can be achieved by: First, obtaining the user's desired outfit for the preset digital avatar. This means that if the user is not satisfied with the outfit of the currently displayed preset digital avatar, they can express their desired outfit through a voice input box or by inputting their desired outfit through a text input box, such as "give the digital avatar a pink formal suit". The user's desired outfit can then be obtained by performing content analysis or intent recognition on the user's voice or text input through a neural network model.

[0064] Step S20: Determine the target outfit model that meets the outfit requirements from the preset outfit database;

[0065] Secondly, a target outfit model that meets the dressing requirements can be obtained by querying a preset outfit database. The preset outfit database can be stored separately based on the ID (Identity document) of different outfit types. The query can be performed by identifying the outfit type ID corresponding to the user's dressing requirements.

[0066] Specifically, a user's dressing needs include the types of clothing and colors they require. For example, the user mentioned above requires formal wear in pink.

[0067] To meet user needs, the specific implementation of determining the target outfit model that meets the dressing requirements from the preset outfit database can also be:

[0068] Determine whether a first garment belonging to the specified clothing category exists in the preset outfit database; if the first garment exists, determine whether a second garment with a color belonging to the specified color category exists in the first garment; if the second garment does not exist, change the pixel color of the first garment based on the color category required by the user to obtain a target outfit model that meets the dressing requirements.

[0069] It is understandable that the system can first check whether the preset outfit database contains a first type of clothing. For example, formal clothing includes first types of clothing such as suit jackets, suit pants, shirts, ties, and leather shoes. If the preset outfit database does not contain the first type of clothing, the system can inform the user that there is no outfit that meets their needs. For example, the system can inform the user that "there is no such outfit, please select another outfit".

[0070] If the first garment mentioned above exists in the preset outfit database, it can be further determined whether there is a second garment in the first garment whose color belongs to the color category. For example, it can be further determined whether the above-mentioned suit jacket, suit pants, shirt, tie, leather shoes, etc. are pink; if they are pink, then the outfit can be changed directly based on the target outfit model.

[0071] If it is not pink, a copy of the corresponding clothing can be made, and the pixel color of the clothing can be changed to the color type desired by the user. For example, if there is no pink shirt but there is a white shirt, a copy of the white shirt can be made, and the pixel color of the white shirt can be changed to pink to obtain the target outfit model that meets the dressing requirements. The pink shirt can be added to the preset outfit database for reuse next time.

[0072] Furthermore, if the user is not satisfied or wants to continue changing outfits, the above steps are repeated until the user no longer enters any new outfit intentions.

[0073] Step S30: Integrate the target costume model and the preset digital human body structure model to obtain the integrated model.

[0074] Specifically, after obtaining the target costume model that meets the costume requirements, the target costume model and the preset digital human body structure model can be integrated to obtain the integrated model.

[0075] One specific implementation method for integrating the target costume model and the preset digital human body structure model is as follows: import the target costume model and the preset digital human body structure model, and integrate the target costume model and the preset digital human body structure model using Blender or Maya (both are open-source 3D creation suites).

[0076] It is understandable that during the costume change process, the human body structure model of the preset digital human remains unchanged. It is only necessary to obtain the target costume model that meets the costume change requirements from the preset costume database, without having to reshoot according to the user's needs.

[0077] Step S30: Based on the integrated model, create the target digital human after the makeover.

[0078] Furthermore, digital humans can be created on the digital human configuration platform. Specifically, a unique digital human identifier can be obtained, and the business, enterprise entity, facial expressions, body movements, lip movements, costume content, and RGB (a color standard representing a combination of red, green, and blue) values ​​of the digital human to be created can be configured. The costume movement trajectory of the new costume model in the integrated model can be calculated. Finally, by generating the corresponding facial expressions, body movements, and lip movements of the digital human (including costume) based on the costume movement trajectory and the corresponding voice according to the configured or real-time generated text, the audio and video are synthesized together and pushed to the digital human interaction terminal for users to watch.

[0079] It is understandable that if the user has already changed the digital persona's clothes, then the default digital persona here is the changed digital persona.

[0080] In this embodiment, the above-described method can meet users' personalized dressing-up needs. Furthermore, in digital human interaction scenarios and in the context of digital humans interacting via audio and video streams, users can change outfits through voice communication, improving the ease of dressing up digital humans. Moreover, it eliminates the need to film a large number of real people, saving costs. Assuming the production cycle for one 2D digital human image (one outfit) is 4-6 weeks and the cost is 300,000, and the production cycle for one 3D digital human image (one outfit) is 4-6 weeks and the cost is 750,000, taking 10 outfits for one digital human as an example, the above method can save 9 production cycles (36-54 weeks), saving 2.7 million for 2D digital humans and 6.75 million for 3D digital humans. Furthermore, unlike related technologies where filming multiple sets of real models in different outfits only yields a limited number of digital humans, this embodiment supports combinations of different outfits and colors from the outfit database, greatly expanding the diversity of outfits and better meeting personalized needs.

[0081] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 Before step S10, the digital human costume changing method further includes steps S01 to S04:

[0082] Step S01: Obtain the original footage obtained by filming real people, the original footage including videos and / or multiple consecutive images;

[0083] It should be noted that in order to obtain the preset digital human based on the mutually separate costume model and human body structure model, the original materials obtained by shooting real people can be obtained first, including videos and / or multiple consecutive pictures.

[0084] Specifically, it could be taking multiple photos of a person in motion wearing the same outfit, or taking multiple videos of a person in motion wearing different outfits.

[0085] Step S02: Extract historical costumes that are separated from the human body structure of real people from the original materials;

[0086] It is understood that by extracting the historical costumes separated from the human body structure of the real person from the original material, the separated historical costumes and the human body structure of the real person can be obtained; specifically, the historical costumes separated from the human body structure of the real person can be extracted from the original material based on image segmentation algorithms, edge detection algorithms, or object recognition algorithms.

[0087] Specifically, the extracted historical costumes can be numbered as c1, c2, c3...cn; the historical costumes can be mapped to the digital human to be created to obtain the mapping relationship between the costumes and the digital human.

[0088] It should be noted that after the step of extracting the historical costume separated from the human body structure of the real person from the original material, it is also possible to: identify the pixel color of the historical costume; determine the number of pixels corresponding to different pixel colors in the historical costume; and color-label the historical costume based on the number of pixels and a preset number threshold to obtain the original color of the historical costume.

[0089] It is understandable that, since different outfits have different colors, the original colors of historical outfits can be determined for easier subsequent management. Specifically, the pixel colors of the historical outfits can be identified, for example, by sequentially extracting the pixel colors of each historical outfit c1, c2, c3...cn; furthermore, the number of pixels corresponding to different pixel colors in the historical outfits can be counted, for example, by determining the number of pixels corresponding to red and green in each historical outfit, thereby obtaining the color set of each historical outfit.

[0090] Furthermore, based on the number of pixels and a preset threshold, the historical outfits can be color-labeled to obtain the original color of the historical outfits. Specifically, a threshold can be preset, for example, setting more than or equal to two colors to consider the historical outfit as a colored garment, and if there is only one color, the historical outfit is considered as a corresponding garment. The original color of the historical outfits can be obtained by color-labeling each historical outfit or by assigning a color label to each historical outfit.

[0091] Because different outfits may overlap—for example, hair and top, pants and shoes—to prevent confusion during the dressing process, after the step of extracting the historical outfits separated from the real human anatomy from the original material, the following can also be done:

[0092] Based on the pixel coordinates of the regions where different historical outfits are located, determine whether there are overlapping regions between the historical outfits; identify the true color of the overlapping regions; based on the true color and the original colors of at least two outfits with overlapping regions, determine the layering relationship between the at least two outfits in the overlapping regions.

[0093] Specifically, since the pixel coordinates of the overlapping areas of two outfits are the same, it is possible to determine whether there are overlapping areas between different historical outfits based on the pixel coordinates of the areas where different historical outfits are located. For example, if the pixel coordinates of at least two outfits are partially the same, it is considered that there is a layering relationship between at least two outfits.

[0094] For example, we can obtain a list Clist consisting of various outfits c1, c2, and c3; take out the i-th (i=1, i.e., the first list element) outfit from the list Clist in sequence, and determine whether the i-th outfit has any overlapping area with other outfits in the list.

[0095] Taking the i-th outfit and the (i+1)-th outfit as examples:

[0096] You can create a hash table (Hashset) and a set of overlapping pixels (CSet). Take out the pixels in the region containing the i-th outfit, concatenate the pixel coordinates (x, y) into a number string (xy), and add xy to the Hashset. Then take out the pixels in the region containing the (i+1)-th outfit, concatenate the pixel coordinates (x, y) into a number string (xy), and check if xy exists in the Hashset. If it does, the two outfits are considered to overlap, and the overlapping point is added to the CSet.

[0097] Furthermore, identifying the true color of the overlapping area can be done specifically by determining the true color of the overlapping area based on the number of pixels in the overlapping area and a preset threshold. The true color of the overlapping area is then compared with the original colors of the two outfits in the overlapping area. If the true color of the overlapping area matches the color of one of the outfits in the overlapping area, then that outfit is considered to be on the upper layer and the other outfit is on the lower layer. For example, when the ends of the hair overlap with the shoulders of the top, the hair outfit is determined to be on the top layer, and the top is on the lower layer in this overlapping area.

[0098] Specifically, based on the pixels in CSet, the true color Color1 of the overlapping area can be identified, and the original colors Color2 and Color3 of the identified i-th and (i+1)-th outfits can be obtained. If Color1 = Color2, then the i-th outfit is considered to be on top of the (i+1)-th outfit; otherwise, if Color1 = Color3, then the (i+1)-th outfit is considered to be on top of the i-th outfit.

[0099] Furthermore, the stacking relationship between the i-th outfit and the (i+1)-th outfit can be recorded in a hash set HashMap (pre-defined stacking relationship library), where the key is in the form of the upper outfit's id minus the lower outfit's id, and the value is the overlapping area of ​​the i-th outfit and the (i+1)-th outfit.

[0100] Since the costume moves during human movement, after the step of extracting the historical costume separated from the real human body structure from the original material, it is also possible to:

[0101] Obtain the first position information of the pixel corresponding to the historical outfit in the video and / or multiple consecutive images, and determine the target outfit type that is moving based on the first position information; fit a motion curve based on the second position information of the pixel corresponding to the target outfit type in the video and / or multiple consecutive images; use the motion curve as the preset outfit motion trajectory corresponding to the target outfit type.

[0102] Specifically, based on the aforementioned series of images or videos (videos are essentially visual effects obtained by rapidly playing multiple images), the movement of historical costumes can be identified sequentially. Specifically, historical costumes may have stationary and moving regions. Stationary regions refer to areas where the costume remains in its original position (i.e., pixels) across multiple images or videos. Moving regions refer to areas where the same pixels change position in different images, as shown in the reference [reference needed]. Figure 3 The position of the hair area changed in three consecutive photos.

[0103] The area to be dressed up can be the area where the center line of the dress up is located. Taking the sleeve as an example, the outer contour of the sleeve can be determined first, and then feature points can be selected. For example, the feature points can be determined as multiple pixels of the outer contour of the sleeve (e.g., cuff, top of the sleeve, center point of the sleeve, etc.). Based on the pixel coordinates of the feature points and the curve corresponding to the outer contour of the sleeve, the center line of the sleeve can be determined.

[0104] Therefore, by obtaining the first position information (e.g., pixel coordinates) of the corresponding pixel in the video and / or multiple consecutive images, the target outfit type that is moving can be determined based on the first position information. For example, if the historical outfit is sleeves, then the corresponding target outfit type can be sleeves. It can be understood that short sleeves, long sleeves, sleeves of tight-fitting clothes, sleeves of gauze skirts, etc. all belong to the outfit type of sleeves.

[0105] Because the movement trajectories within the same area are highly similar—for example, the swing curves of a real person's arm are often very similar—based on this fact, and also to reduce computational load, the movement trajectory can be calculated based on the set of pixels in the middle of the area. Specifically, the position coordinates of pixels in the same area can be recorded in multiple images (with the lower left corner of the image as the origin of the two-dimensional coordinate axis, the x-axis to the right, and the y-axis upward). A curve can be fitted and calculated based on the coordinates of these points; this curve represents the movement trajectory of the corresponding area of ​​the costume. Figure 4This allows us to obtain the movement trajectory of the hair.

[0106] Furthermore, when it is necessary to extend the movement trajectory of the costume, for example, when the sleeves of the costume change become longer or shorter, the trajectory of the shortened sleeve or the lengthened sleeve can be deduced based on the above-mentioned movement trajectory. Specifically, based on the knowledge of biomechanics, the direction of motion and force characteristics of the current movement trajectory of the costume can be analyzed to predict the trajectory of the shortened sleeve or the lengthened sleeve, thereby achieving coordination between human movement and costume movement.

[0107] Step S03: Model the historical outfit to obtain the historical outfit model;

[0108] Furthermore, the historical costume can be modeled to obtain a historical costume model, and / or the extracted human body structure can be modeled to obtain a preset human body structure model. It can be understood that this embodiment models the historical costume and human body structure separately, which is different from the traditional modeling method that combines them into one.

[0109] Step S04: Based on the historical costume model and the preset human body structure model, create a preset digital human.

[0110] It should be noted that the preset human body structure model can be based on the human body structure model extracted from the historical material, or it can be based on other human body structures in advance.

[0111] Furthermore, the historical costume model and the preset human body structure model can be integrated to obtain an integrated historical model. Specifically, before integrating the historical costume model and the preset human body structure model, facial expression modeling, body movement modeling, lip movement modeling, and AI training can be performed, and then integrated with the historical costume model and the preset human body structure model respectively; then, based on the integrated model, a preset digital human is created.

[0112] In this embodiment, the original colors of each outfit, the layering relationships between each outfit, and the preset movement trajectories of each outfit are obtained through the above methods. This facilitates the integration of the target outfit model and the human body structure model based on the original colors and layering relationships during the dress-up process. It also facilitates the setting of the digital human's movement trajectory based on the preset movement trajectories of each outfit during the digital human creation process, thereby serving the dress-up process and improving the accuracy of digital human creation.

[0113] Based on the first and second embodiments of this application, the content that is the same as or similar to that in Embodiment 1 above in this third embodiment can be referred to the above description and will not be repeated hereafter. Based on this, please refer to... Figure 5 Step S30 includes steps A1 to A3:

[0114] Step A1: Based on a preset layering relationship library, determine whether the target costume model overlaps with the original costume model of the preset digital human. The preset layering relationship library stores the layering relationships between different costume models.

[0115] It should be noted that when there is an overlapping area between the target costume model ci and the surrounding costume cj, the layering relationship between the two needs to be processed. Therefore, based on the preset layering relationship library, it can be determined whether the target costume model has an overlapping area with the original costume model of the preset digital human.

[0116] Specifically, it can be done by checking if the target costume model ci exists in the keys of the hash collection HashMap. If it does not exist, it is assumed that the target costume model has no cascading relationship with any costume, and no cascading relationship processing is required.

[0117] Step A2: If it exists, then based on the layering relationship between the target outfit model and the original outfit model, and the preset original colors of the target outfit model and the original outfit model at the overlapping area, determine the target color at the overlapping area after the outfit change;

[0118] If it exists, then query the key-value pair of the outfit ci's id, where the key is in the form of "the id of the upper outfit - the id of the lower outfit". Therefore, based on the key, the outfit id that overlaps with ci and the corresponding stacking relationship can be obtained. That is, the stacking relationship between the target outfit model and the original outfit model can be obtained.

[0119] Furthermore, the color of the overlapping area can be obtained by individually determining the coordinates and stacking relationship of the overlapping area in the corresponding value of the HashMap. At this point, the color of the overlapping area can be changed to the color of the upper layer decoration.

[0120] Step A3: Determine a new outfit model based on the target outfit model, the original outfit model of the preset digital human, and the target color;

[0121] Furthermore, based on the target color, the target costume model and the original costume model of the preset digital human can be integrated to obtain a new costume model. It can be understood that color processing based on the layering relationship is required during the integration process.

[0122] Step A4: Integrate the new costume model and the preset digital human body structure model to obtain the integrated model.

[0123] Furthermore, by integrating the new costume model and the preset digital human body structure model, an integrated model can be obtained, and the target digital human after the costume change can be created based on the integrated model.

[0124] Specifically, the specific implementation method for creating the target digital human after the makeover based on the integrated model can be:

[0125] Based on the target outfit type corresponding to the new outfit model and the preset outfit movement trajectory corresponding to different outfit types, the target outfit movement trajectory of the new outfit model is determined; based on the integrated model and the target outfit movement trajectory, the target digital human after the outfit change is created.

[0126] It should be noted that after the costume change, since some or all of the costumes of the preset digital human have changed, the corresponding preset costume movement trajectory may also change. Specifically, the target costume movement trajectory of the new costume model can be determined based on the target costume type corresponding to the new costume model and the preset costume movement trajectories corresponding to different costume types.

[0127] For example, if the target outfit type corresponding to the new outfit model is sleeves, and there is a predetermined preset outfit motion trajectory for sleeves, then the preset outfit motion trajectory corresponding to the sleeves can be used as the target outfit motion trajectory for the new outfit model.

[0128] The target outfit type is the type of outfit that moves when the human body moves (the type of outfit corresponds to the area of ​​movement) as determined above.

[0129] In this embodiment, the target costume model and human body structure model are integrated based on the original colors and layering relationships during the digital human costume change process. The motion trajectory of the digital human is set based on the preset motion trajectory of each costume, which improves the accuracy of digital human creation and costume change efficiency.

[0130] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the digital human costume-changing method of this application. Any simple transformations based on this technical concept are within the protection scope of this application.

[0131] This application provides a digital human costume changing device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the digital human costume changing method in the first embodiment described above.

[0132] The following is for reference. Figure 6The diagram illustrates a structural schematic suitable for implementing the digital human costume-changing device in the embodiments of this application. The digital human costume-changing device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, tablets, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 6 The digital human dressing device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0133] like Figure 6 As shown, the digital human costume-changing device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the digital human costume-changing device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the digital human costume change device to communicate wirelessly or wiredly with other devices to exchange data. Although the figures show digital human costume change devices with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0134] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0135] The digital human costume-changing device provided in this application, employing the digital human costume-changing method described in the above embodiments, can solve the technical problem of how to reduce the cost of digital human costume-changing while meeting users' personalized needs. Compared with the prior art, the beneficial effects of the digital human costume-changing device provided in this application are the same as those of the digital human costume-changing method described in the above embodiments, and other technical features of this digital human costume-changing device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0136] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0137] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0138] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the digital human dressing method described in the above embodiments.

[0139] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0140] The aforementioned computer-readable storage medium may be included in the digital human costume changing device; or it may exist independently and not be installed in the digital human costume changing device.

[0141] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the digital human costume device, cause the digital human costume device to perform the aforementioned digital human costume method.

[0142] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can 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 can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0143] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0144] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0145] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described digital human costume-changing method. This solves the technical problem of reducing the cost of digital human costume-changing while meeting users' personalized needs. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the digital human costume-changing method provided in the above embodiments, and will not be repeated here.

[0146] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the digital human costume-changing method described above.

[0147] The computer program product provided in this application can solve the technical problem of reducing the cost of digital human costume changing while meeting users' personalized needs. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the digital human costume changing method provided in the above embodiments, and will not be repeated here.

[0148] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for digital human costume changing, characterized in that, The method includes: The system obtains the user's clothing requirements for a preset digital human, wherein the preset digital human is created based on a separate clothing model and a human body structure model. Determine the target outfit model that meets the dressing requirements from the preset outfit database; The target costume model and the preset digital human body structure model are integrated to obtain the integrated model; Based on the integrated model, the target digital human after the makeover is created.

2. The method as described in claim 1, characterized in that, The step of integrating the target costume model and the preset digital human body structure model to obtain the integrated model includes: Based on a preset layering relationship library, it is determined whether the target costume model overlaps with the original costume model of the preset digital human. The preset layering relationship library stores the layering relationships between different costume models. If it exists, then based on the layering relationship between the target outfit model and the original outfit model, and the preset original colors of the target outfit model and the original outfit model at the overlapping area, the target color at the overlapping area after the outfit change is determined; Based on the target costume model, the original costume model of the preset digital human, and the target color, a new costume model is determined; The new costume model and the preset digital human body structure model are integrated to obtain the integrated model.

3. The method as described in claim 2, characterized in that, The step of creating the target digital human after the makeover based on the integrated model includes: Based on the target outfit type corresponding to the new outfit model and the preset outfit movement trajectory corresponding to different outfit types, the target outfit movement trajectory of the new outfit model is determined; Based on the integrated model and the target costume's movement trajectory, a target digital human after costume change is created.

4. The method as described in claim 1, characterized in that, Before the step of obtaining the user's clothing requirements for the preset digital human, the method further includes: Obtain raw footage of real people, including videos and / or multiple consecutive images; Historical costumes, separated from the human anatomy of real people, were extracted from the original materials. The historical costume is modeled to obtain the historical costume model; Based on the historical costume model and the preset human body structure model, a preset digital human is created.

5. The method as described in claim 4, characterized in that, After the step of extracting historical costumes that are separated from the human anatomy of real people from the original materials, the method further includes: Identify the pixel colors of the historical outfits; Determine the number of pixels corresponding to different pixel colors in the historical outfit; Based on the number of pixels and a preset threshold, the historical outfit is color-coded to obtain the original color of the historical outfit.

6. The method as described in claim 5, characterized in that, After the step of extracting historical costumes that are separated from the human anatomy of real people from the original materials, the method further includes: Based on the pixel coordinates of the regions where different historical costumes are located, determine whether there are overlapping areas between the historical costumes; Identify the true color of overlapping areas; Based on the actual color and the original colors of at least two outfits that have overlapping areas, determine the layering relationship between the at least two outfits in the overlapping areas.

7. The method as described in claim 4, characterized in that, After the step of extracting historical costumes that are separated from the human anatomy of real people from the original materials, the method further includes: Obtain the first position information of the pixel corresponding to the historical outfit in the video and / or multiple consecutive images, and determine the target outfit type that is moving based on the first position information; Based on the second position information of the pixels corresponding to the target outfit type in the video and / or multiple consecutive images, a motion curve is fitted to obtain the motion curve. The motion curve is used as the preset motion trajectory of the target outfit type.

8. A digital human dressing device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the digital human dressing method as described in any one of claims 1 to 7.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the digital human dressing method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the digital human dressing method as described in any one of claims 1 to 7.

Images