WO2021143278A1 - 图像处理方法、装置、电子设备及存储介质 - Google Patents

图像处理方法、装置、电子设备及存储介质 Download PDF

Info

Publication number
WO2021143278A1
WO2021143278A1 PCT/CN2020/125081 CN2020125081W WO2021143278A1 WO 2021143278 A1 WO2021143278 A1 WO 2021143278A1 CN 2020125081 W CN2020125081 W CN 2020125081W WO 2021143278 A1 WO2021143278 A1 WO 2021143278A1
Authority
WO
WIPO (PCT)
Prior art keywords
bone
surface model
posture
pose
model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2020/125081
Other languages
English (en)
French (fr)
Inventor
刘杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to JP2022520106A priority Critical patent/JP7299414B2/ja
Priority to EP20913250.5A priority patent/EP4002286A4/en
Priority to KR1020227003382A priority patent/KR102698917B1/ko
Publication of WO2021143278A1 publication Critical patent/WO2021143278A1/zh
Priority to US17/706,242 priority patent/US12183036B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating three-dimensional [3D] models or images for computer graphics
    • G06T19/20—Editing of three-dimensional [3D] images, e.g. changing shapes or colours, aligning objects or positioning parts
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
    • A63F13/30—Interconnection arrangements between game servers and game devices; Interconnection arrangements between game devices; Interconnection arrangements between game servers
    • A63F13/35—Details of game servers
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
    • A63F13/55—Controlling game characters or game objects based on the game progress
    • A63F13/56—Computing the motion of game characters with respect to other game characters, game objects or elements of the game scene, e.g. for simulating the behaviour of a group of virtual soldiers or for path finding
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
    • A63F13/60—Generating or modifying game content before or while executing the game program, e.g. authoring tools specially adapted for game development or game-integrated level editor
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T13/00—Animation
    • G06T13/20—Three-dimensional [3D] animation
    • G06T13/40—Three-dimensional [3D] animation of characters, e.g. humans, animals or virtual beings
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three-dimensional [3D] modelling for computer graphics
    • G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating three-dimensional [3D] models or images for computer graphics
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • G06V40/23—Recognition of whole body movements, e.g. for sport training
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F2300/00—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
    • A63F2300/60—Methods for processing data by generating or executing the game program
    • A63F2300/66—Methods for processing data by generating or executing the game program for rendering three dimensional images
    • A63F2300/6607—Methods for processing data by generating or executing the game program for rendering three dimensional images for animating game characters, e.g. skeleton kinematics
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
    • G06T2219/20—Indexing scheme for editing of 3D models
    • G06T2219/2021—Shape modification

Definitions

  • This application relates to the field of image processing, in particular to an image processing method, device, electronic equipment, and computer-readable storage medium.
  • the embodiment of the present application provides an image processing method, including:
  • the example surface model and the reference surface model are converted to the same coordinate space, so as to perform the virtual object's calculation according to the reference surface model and the example surface model in the coordinate space. generate.
  • An embodiment of the application provides an image processing device, including:
  • the first acquiring module is configured to acquire the reference surface model of the first bone pose of the virtual object
  • the second acquiring module is configured to acquire the second bone posture of the virtual object
  • a generating module configured to generate an example surface model of the second bone pose based on the reference surface model and the second bone pose
  • a matrix determination module configured to determine a posture transformation matrix between the first bone posture and the second bone posture
  • the coordinate space conversion module is configured to convert the example surface model and the reference surface model to the same coordinate space based on the pose transformation matrix, so as to according to the reference surface model and the example surface in the coordinate space
  • the model generates the virtual object.
  • An embodiment of the present application provides an electronic device, including: a memory configured to store computer-readable instructions; a processor configured to execute the computer-readable instructions stored in the memory to implement the Image processing method.
  • the embodiment of the present application provides a computer-readable storage medium that stores computer-readable instructions.
  • the computer-readable instruction is executed by a processor, the image processing method provided in the embodiment of the present application is implemented.
  • FIG. 1 shows a schematic diagram of the architecture of an image processing system provided by an embodiment of the present application.
  • Fig. 2 shows a schematic structural diagram of an image processing system applied to a game scene provided by an embodiment of the present application.
  • Fig. 3A shows a flowchart of an image processing method provided by an embodiment of the present application.
  • Fig. 3B shows a flowchart of an image processing method provided by an embodiment of the present application.
  • Fig. 3C shows a flowchart of an image processing method provided by an embodiment of the present application.
  • Fig. 4 shows a flow chart of the material resource generation stage provided by an embodiment of the present application.
  • Fig. 5 shows a flowchart of the virtual object generation phase provided by an embodiment of the present application.
  • Fig. 6 shows a terminal interface diagram for generating virtual objects in Maya software provided by an embodiment of the present application.
  • Fig. 7 shows a block diagram of an image processing device provided by an embodiment of the present application.
  • Fig. 8 shows a hardware diagram of an electronic device provided by an embodiment of the present application.
  • Virtual scene The scene output by the device is different from the real world, and the visual perception of the virtual scene can be formed with the aid of the naked eye or the device, such as the two-dimensional image output through the display screen, through stereo projection, virtual reality and augmented reality.
  • Virtual objects the images of various people and things that can interact in the virtual scene, or the movable objects in the virtual scene, which can be generated by game engines, drawing software or other image processing tools to generate virtual objects
  • the core technology is three-dimensional image technology.
  • the movable object may be a virtual character, a virtual animal, an animation character, etc., for example, a character, an animal, a plant, an oil barrel, a wall, a stone, etc. displayed in a virtual scene.
  • the virtual object may be a virtual avatar used to represent the user in the virtual scene.
  • the virtual object has its own shape and volume in the virtual scene and occupies a part of the space in the virtual scene.
  • the virtual object may be an animated character, and the image processing method provided in the embodiment of the present application may be applicable to the image processing process of the animated character in skeletal animation.
  • Skeletal Animation It is a kind of model animation.
  • virtual objects such as animated characters
  • the bone posture describes the orientation and position of the "skeleton" of the virtual object
  • the surface model describes multiple vertices on the surface of the virtual object, that is, the surface model can be understood as a collection of a series of vertices on the surface of the virtual object.
  • the change of the bone posture can correspondingly drive the change of the surface model, so that the corresponding animation can be generated by driving the bone posture.
  • Skeleton skinning model refers to a model with bone skinning information after skinning the surface model according to the bone posture.
  • the bone skin model mainly includes two parts of information: surface information (surface data) and bone skin information.
  • the material resources include a reference surface model and a sample surface model.
  • the reference surface model is the most basic preset material resource, and there can be only one;
  • the example surface model is a material resource further obtained on the basis of the reference surface model, and there can be only one or at least two.
  • the embodiments of the present application provide an image processing method, device, electronic device, and computer-readable storage medium, which can skip the effect of surface deformation in the process of producing material resources, thereby supporting the generation of virtual objects.
  • Surface deformation treatment improves the compatibility with surface deformation.
  • FIG. 1 shows a schematic structural diagram of an image processing system provided by an embodiment of the present application.
  • the image processing system includes a first terminal device 10, a server 20 and a second terminal device 30.
  • the image processing method provided in the embodiments of the present application may be implemented by the terminal device.
  • the first terminal device 10 is taken as an example for description.
  • the first terminal device 10 may perform the operation of generating material resources.
  • the first terminal device 10 may provide the first user with functions related to making material resources, such as the function of determining the surface model, the function of adjusting the bone posture, etc. Generate material resources under the operation of the first user.
  • the first terminal device 10 can store the generated material resources locally, and when necessary, generate virtual objects based on the stored material resources, and render the generated virtual objects to display the virtual objects, such as the output Virtual objects are displayed in the virtual scene. This method is suitable for scenarios where the terminal device has a strong computing capability.
  • the image processing method provided in the embodiments of the present application may be implemented by the server.
  • the server 20 may perform automatic generation of material resources, for example, generate material resources according to pre-stored first bone pose, second bone pose, and reference surface model, and store the generated material resources locally.
  • the server 20 may also generate a virtual object according to the stored material resources, and send the generated rendering data of the virtual object to the terminal device, so that the terminal device displays the virtual object, such as displaying the virtual object in the virtual scene output by the terminal device.
  • the image processing method provided in the embodiments of the present application may also be implemented by a terminal device and a server in cooperation.
  • the first terminal device 10 and the server 20 are mainly involved.
  • the first terminal device 10 is mainly used as a front end to respond to the operation of the first user to generate material resources;
  • the server 20 is mainly used as a back end to provide corresponding functions for the first terminal device 10 in the process of generating material resources Support (such as the function of determining the surface model, the function of adjusting the bone posture, etc.), and the material resources sent by the first terminal device 10 are stored.
  • the second terminal device 30 and the server 20 are mainly involved.
  • the server 20 is mainly used as a back-end for generating virtual objects based on stored material resources, and sending the generated rendering data of the virtual objects to the second terminal device 30;
  • the second terminal device 30 is mainly used as a front-end Rendering is performed according to the rendering data received from the server 20 to display the virtual object, such as display in the output virtual scene, to facilitate viewing or interaction by the second user.
  • FIG. 1 shows a virtual scene 31 output by the second terminal device 30 and a virtual object 32 in the virtual scene 31.
  • the virtual scene 31 and the virtual object 32 can be displayed on the graphical interface of the second terminal device 30. .
  • Figure 2 shows a schematic diagram of the architecture of the image processing system in a game scenario in an embodiment of the present application.
  • the image processing system in Figure 2 includes a modeler terminal device 40, a game server 50, and a game player terminal.
  • Equipment 60 For ease of understanding, a case where a terminal device and a server cooperate to implement the image processing method provided in the embodiment of the present application is taken as an example for description.
  • the modeler terminal device 40 and the game server 50 are mainly involved.
  • the modeler terminal device 40 is mainly used as a front end to respond to the modeler's production operations on game material resources (for example, the surface model of the game character with a natural standing posture, the surface model of the game character with the open arms, etc.).
  • game material resources for example, the surface model of the game character with a natural standing posture, the surface model of the game character with the open arms, etc.
  • Generate game material resources the game server 50 is mainly used as a backend to provide corresponding functional support for the modeler terminal device 40 during the process of generating the game material resources, and store the game material resources sent by the modeler terminal device 40.
  • the modeler can perform corresponding art work on the modeler terminal device 40 (for example, modify the surface model of the game character through the digital drawing board, modify the surface model of the game character through mouse commands and keyboard commands, etc.).
  • the modeler terminal The device 40 can generate corresponding game material resources with the support of the game server 50.
  • the game player terminal device 60 and the game server 50 are mainly involved.
  • the game server 50 is mainly used as a backend to generate virtual objects based on stored game material resources.
  • the virtual object is a game character as an example.
  • the game server 50 is also used to generate rendering data of the game character. It is sent to the game player terminal device 60; the game player terminal device 60 mainly serves as a front end for rendering according to the rendering data received from the game server 50 to display the game character, such as displaying in the output virtual game scene.
  • FIG. 2 shows a game virtual scene 61 output by the game player terminal device 60 and a game character 62 in the game virtual scene 61.
  • the game player can perform game operations on the game player terminal device 60 (for example, controlling the movement of the game character 62 through the mouse, controlling the skill release of the game character 62 through the keyboard, controlling the movement and skill release of the game character 62 through touch on the screen, etc.),
  • the game character 62 displayed on the game player terminal device 60 is controlled.
  • the game player terminal device 60 generates a corresponding instruction in response to the player's game operation, and sends the instruction to the game server 50.
  • the game server 50 may generate a new game character in combination with the stored game material resources according to the received instruction.
  • the game character 62 shown in FIG. 2 is in a jumping state, and the generated new game character may be a game character in a standing state. 62, that is, the bone posture has changed.
  • the terminal device can implement the image processing method provided in the embodiments of the present application by running a computer program.
  • the computer program can be a native program or a software module in the operating system; it can be a native application program (APP, Application), that is, a program that needs to be installed in the operating system to run, such as a game application; it can also be a small program, that is, a program that only needs to be downloaded to the browser environment to run; it can also be embedded in any Small programs in the APP.
  • APP Native application program
  • the above-mentioned computer program may be any form of application, module or plug-in.
  • Cloud technology refers to a kind of hosting that unifies a series of resources such as hardware, software, and network in a wide area network or a local area network to realize the calculation, storage, processing and sharing of data.
  • cloud technology is also a general term for network technology, information technology, integration technology, management platform technology, and application technology based on the application of cloud computing business models. It can form a resource pool, which can be used on demand and is flexible and convenient. Cloud computing technology will become an important support. The background service of the technical network system requires a lot of computing and storage resources.
  • the above-mentioned server may be an independent physical server, or a server cluster or a distributed system composed of multiple physical servers, or it may provide cloud services, cloud databases, cloud computing, cloud functions, and cloud storage.
  • Cloud servers for basic cloud computing services such as, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.
  • cloud services can be services that generate material resources or generate virtual objects Service for the terminal device to call.
  • the terminal device can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart watch, a smart TV, etc., but it is not limited to this.
  • the terminal device and the server can be directly or indirectly connected through wired or wireless communication, which is not limited in the embodiment of the present application.
  • FIG. 3A is a schematic flowchart of an image processing method provided by an embodiment of the present application, which will be described in conjunction with the steps shown in FIG. 3A.
  • Step 410 Obtain a reference surface model of the first bone pose of the virtual object.
  • Step 420 Obtain a second bone pose of the virtual object.
  • Step 430 Generate an example surface model of the second bone pose based on the reference surface model and the second bone pose;
  • Step 440 Determine a pose transformation matrix between the first bone pose and the second bone pose.
  • Step 450 Based on the attitude transformation matrix, transform the example surface model and the reference surface model to the same coordinate space, so as to generate a virtual object according to the reference surface model and the example surface model in the coordinate space.
  • the electronic device in the generation stage of the material resources, after the electronic device generates the sample surface model according to the reference surface model, it is based on the first bone pose (the bone pose corresponding to the reference surface model) and the second bone pose (the sample surface model corresponds to The posture transformation matrix between the skeletal postures) converts the sample surface model and the reference surface model to the same coordinate space to obtain the material resources in the same coordinate space.
  • the material resource generation process can skip the influence of additional surface deformation, so that the subsequent virtual object generation stage can be compatible with additional surface deformations (such as various complex nonlinear additional surface deformations), and can effectively avoid different
  • additional surface deformations such as various complex nonlinear additional surface deformations
  • the result of processing in the coordinate space is uncontrollable (for example, the final generated virtual object does not match the applied surface deformation), which improves the compatibility of the skeletal animation to the surface deformation.
  • step 410 a reference surface model of the first bone pose of the virtual object is acquired.
  • virtual objects refer to various objects of people and things that are virtualized.
  • virtual objects can be animated characters, trees, flags, or paper.
  • the virtual object has a variety of skeletal postures.
  • the animated character has a skeleton posture in a natural standing posture and a skeleton posture in a horizontal posture with arms.
  • the flag has a skeleton posture when it is stationary and when it is blown by the wind. The posture of the skeleton and so on.
  • the surface model of the first bone pose of the virtual object is obtained as the reference surface model.
  • the reference surface model may be the surface model of the virtual object in the natural stance
  • the first bone posture is the skeletal posture of the virtual object in the natural stance.
  • the reference surface model can be generated in advance and stored in the electronic device.
  • the electronic device can retrieve the reference surface model; the reference surface model can also be used by the user (for example, The modeler who is doing art work will make it in real time on the electronic device (such as the modeler's terminal device), and the same is true for the first skeleton posture.
  • step 420 the second bone pose of the virtual object is acquired.
  • the second bone pose refers to the bone pose of the sample surface model to be generated; the sample surface model refers to the surface model that is generated with the reference surface model as a reference and is not the first bone pose.
  • the reference surface model is the surface model of the virtual object in the natural stance
  • the surface model of the virtual object in the natural stance is used as a reference.
  • the generated surface model of the virtual object in the horizontal posture of both arms is an example surface.
  • Model the skeleton posture of the virtual object in the horizontal posture of both arms is the second skeleton posture.
  • the second skeleton posture can be generated in advance and stored in the electronic device; it can also be created on the electronic device in real time by the user (for example, a modeler who performs art work).
  • the embodiment of the present application does not limit the number of the second bone postures, for example, it may be one or at least two.
  • FIG. 3A illustrates the case where the second bone posture includes only one.
  • step 430 based on the reference surface model and the second bone pose, an example surface model of the second bone pose is generated.
  • a surface model of the second bone pose can be generated.
  • the surface model generated here is named an example surface model.
  • the above-mentioned example surface model based on the reference surface model and the second bone pose can be realized in this way: the reference surface model is skinned based on the second bone pose to obtain The example bone skin model of the second bone pose; extract the surface data of the example bone skin model, and generate the example surface model of the second bone pose according to the surface data.
  • the bone skin model mainly includes two parts of information: surface data and bone skin information.
  • the bone skin model obtained here is named as the model bone skin model.
  • the reference surface model is a horizontal arm model in a horizontal posture
  • the second bone posture is a skeletal posture of an upward arm in an upward posture.
  • the horizontal arm model is skinned based on the skeletal posture of the upward arm to obtain the upward arm model with the skeletal skin information and the upward posture, that is, the example skeletal skin model.
  • the example bone skin model of the second bone pose After obtaining the example bone skin model of the second bone pose, extract its surface data, and then generate the example surface model of the second bone pose based on the surface data.
  • the example surface model includes the virtual model described by the extracted surface data. The vertices on the surface of the object.
  • the method before extracting the surface data of the example skeleton skin model, the method further includes:
  • the above-mentioned extraction of the surface data of the example bone skin model can be achieved in this way: the surface data of the modified example bone skin model is extracted.
  • the requesting party (for example, at the front end)
  • the modeler's terminal equipment) modify the example skeleton skin model.
  • the skinning process itself is a linear mathematical process. Therefore, the sample bone skin model obtained by the skin process only makes a simple bone posture change.
  • the surface data in the sample bone skin model may not meet the actual needs.
  • a curved arm model (skeletal skin model) with the elbow bending posture with bone skin information can be obtained.
  • the surface of the biceps in the horizontal arm model is relatively flat, and the surface of the biceps in the curved arm model obtained after skin treatment hardly changes significantly, and is still relatively flat. It is understandable that in real life, after the elbow is bent, it will cause a backlog of muscles. Compared with the horizontal arm model, the surface of the biceps in the bent arm model should be significantly raised. If there is a specific art requirement, the bulge of the surface of the biceps in the flexed arm model will be more obvious.
  • the requesting party can be supported to modify the example skinned skeleton model to meet corresponding needs, such as art needs.
  • the user sends a modification request for the sample skeleton skin model to an electronic device (such as a server) through a request.
  • an electronic device such as a server
  • the electronic device passes the verification of the received modification request
  • the corresponding modification authority is opened to the requester, so that the user can The requester modifies the example skeleton skin model.
  • the modeler sends a modification request to the game server through the modeler's terminal device according to artistic needs, so as to request to modify the example skeleton skin model obtained through the skinning process.
  • the game server passes the verification of the modification request, it will open the corresponding modification authority to the modeler's terminal device. In this way, the modeler can modify the example skeleton skin model on the modeler's terminal device.
  • the user can modify the surface data in the example bone skin model during the requesting party's modification of the example bone skin model; or modify the second bone pose corresponding to the example bone skin model to update the example bone skin model Surface data in.
  • the user can send the modified message to the electronic device through the requester.
  • the electronic device receives the modified message, it extracts the surface data in the modified example bone skin model to generate it based on the surface data.
  • Example surface model of the second bone pose Through the above-mentioned method of combining with human modification, the accuracy of the surface data in the example skeleton skin model can be improved, and various requirements in actual application scenarios can be met.
  • step 440 a pose transformation matrix between the first bone pose and the second bone pose is determined.
  • the posture transformation matrix refers to a mathematical matrix used to describe the mutual transformation relationship between the first bone posture and the second bone posture. Through the posture transformation matrix, the first bone posture can be transformed into the second bone posture, and the second bone posture can also be transformed into the first bone posture.
  • the first bone posture and the second bone posture are both in the form of specific data in the electronic device.
  • the electronic device can store the three-dimensional space coordinates of the bone vertices, and by adjusting the three-dimensional space coordinates of the bone vertices, the corresponding bone posture can be obtained, that is, the bone posture can be stored and processed in the form of three-dimensional space coordinate data.
  • step 450 based on the pose transformation matrix, the example surface model and the reference surface model are converted to the same coordinate space, so as to generate a virtual object according to the reference surface model and the example surface model in the coordinate space.
  • the generated material resources are converted into the same coordinate space in order to generate virtual objects. That is, in the generation stage of the material resources, the sample surface model and the reference surface model are converted to the same coordinate space.
  • the example surface model can be converted to the coordinate space where the reference surface model is located, or the reference surface model can be converted to the coordinate space where the example surface model is located.
  • the above-mentioned posture transformation matrix can be used to transform the sample surface model and the reference surface model into the same coordinate space: according to the posture transformation matrix, coordinate transformation processing is performed on the sample surface model to transform the sample surface model into the same coordinate space. The sample surface model is converted to the coordinate space where the reference surface model is located.
  • the reference surface model is a standard set by humans. Therefore, coordinate transformation processing can be performed on the example surface model according to the pose transformation matrix to convert the example surface model to the coordinate space where the reference surface model is located. Furthermore, the reference surface model and the example surface model can be stored in the coordinate space where the reference surface model is located to complete the generation of material resources.
  • the reference surface model M 0 is located in a coordinate space.
  • the coordinate space is named the reference coordinate space.
  • the electronic device initially acquired data includes a reference surface model M 0, M 0 reference surface model corresponding to a first bone posture P 0, examples of the surface to be generated model M i corresponding to the second bone posture P i.
  • First bone and the second bone posture posture P 0 P i is calculated according to the posture transformation matrix between both, through the attitude transformation matrix transform the coordinate space in a subsequent process.
  • the sample may be further skinned model by the user (e.g., modeler) according to actual needs (like the art) Modify to obtain a new example bone skin model that meets the requirements (that is, the modified example bone skin model).
  • the user e.g., modeler
  • Modify to obtain a new example bone skin model that meets the requirements (that is, the modified example bone skin model).
  • the electronic device stores the reference surface model M 0 and the example surface model M i in the reference coordinate space through the attitude transformation matrix, so that in the subsequent virtual object generation process, the reference surface model M 0 and the example surface model surface model M i series of processes are performed in the reference coordinate space, thereby skipping the influence of the surface deformation.
  • assets generated i.e. the reference surface model surface model M 0 and M i examples be compatible with any additional surface deformation, such as the various additional complex nonlinear surface deformation, to effectively prevent processed in different coordinate space
  • the result is uncontrollable, for example, the final virtual object does not match the applied surface deformation.
  • FIG. 3B is a schematic flowchart of the image processing method provided by an embodiment of the present application. Based on FIG. 3A, after step 450, the target bone pose of the virtual object can be obtained in step 510. .
  • the target bone pose refers to the bone pose of the virtual object that needs to be generated, which can be set according to actual application scenarios.
  • the target skeletal posture is a skeletal posture in a standing posture with hands on hips
  • the final virtual object generated is a standing posture with hands on hips.
  • the target bone posture may be generated in advance and stored in the electronic device, or it may be obtained in real time.
  • the aforementioned acquisition of the target bone posture of the virtual object can be achieved in such a way: obtaining manipulation data for the virtual object; and determining the target bone posture of the virtual object based on the manipulation data.
  • the target bone posture can be determined by the manipulation data for the virtual object, such as manipulation data input based on a specific input device.
  • the manipulation data may originate from the input device in the second terminal device 30 in FIG. 1 or the input device in the game player terminal device 60 in FIG. 2.
  • the manipulation data may include at least one of keyboard data, mouse data, and screen touch data, but of course it is not limited thereto.
  • the player can perform a screen touch operation on the mobile phone terminal, thereby instructing the game character in the game running on the mobile phone terminal to make a corresponding action in a specific posture.
  • the mobile terminal detects the screen touch operation, it uploads the screen touch data triggered by the screen touch operation to the game server, and the game server determines the specific posture indicated by the player based on the acquired screen touch data, that is, the target bone posture of the game character .
  • the user can determine the target bone posture by active manipulation, thereby controlling the generation of the virtual object.
  • the above-mentioned acquisition of the target bone posture of the virtual object can be achieved in such a way: sensor data is obtained; the user action is determined based on the sensor data; the target bone posture of the virtual object is determined based on the user action.
  • the target bone posture of the virtual object can also be determined through sensor data.
  • the sensor data may come from a sensor device, for example, the sensor device in the second terminal device 30 in FIG. 1 or the sensor device in the game player terminal device 60 in FIG. 2.
  • the sensor data may include at least one of positioning sensor data related to positioning information and motion sensor data related to local body movements.
  • the player wears a virtual reality (VR) device that communicates with a computer terminal, such as VR glasses or a VR handle, and makes specific actions to instruct the VR game running on the computer terminal.
  • the corresponding game character makes the same action.
  • the VR sensor system corresponding to the VR device (for example, a sensor system composed of at least two base stations) can generate corresponding sensor data by monitoring the sensors in the VR device (for example, the sensor in the VR handle), and send the sensor data to the computer terminal .
  • the server can determine the user's motion (ie, realize motion capture) based on the sensor data, and use the determined skeletal posture corresponding to the user's action as the target skeletal posture of the game character.
  • the user can determine the target bone posture through his own actions, which can bring a greater sense of immersion to the user and enhance the user's interactive experience.
  • step 520 skinning is performed on the reference surface model based on the target bone pose to obtain the target bone skin model of the target bone pose.
  • the reference surface model is skinned based on the target bone pose to obtain the bone skin model of the target bone pose.
  • the bone skin model obtained here is named the target bone skin model.
  • step 530 based on the reference surface model and the example surface model in the coordinate space, a modified surface model for the posture of the target bone is determined.
  • the modified surface model of the target bone pose is determined.
  • the modified surface model refers to a surface model that can make a certain degree of modification to the surface data in the target bone skin model to meet the actual needs (such as art needs) to a certain extent.
  • the bone pose corresponding to the modified surface model is basically the same as the bone pose corresponding to the target bone skin model, which is the target bone pose.
  • the generated surface model of the target bone pose (that is, the modified surface model) also belongs to the art style. In this way, it can ensure that the generated image based on different target bone poses
  • the style of the virtual objects is consistent, which enhances unity.
  • the above-mentioned reference surface model and example surface model in the coordinate space can be implemented in this way to determine the modified surface model of the target bone pose: it is determined that the first bone pose and the second bone pose are merged.
  • the weights assigned to the first bone pose and the second bone pose respectively are fused to obtain the target Modified surface model of bone pose.
  • the weights assigned to the first bone posture and the second bone posture are determined, wherein, according to the weights assigned to the first bone posture and the second bone posture, the first bone posture and the second bone posture can be fused into the target For bone pose, the fusion here can be a weighted summation, of course, it can also be in other ways. Furthermore, based on the weight of the first bone pose and the weight of the second bone pose, the reference surface model of the first bone pose and the example surface model of the second bone pose are fused to obtain a modified surface model of the target bone pose.
  • the above determination can be achieved in this way.
  • the weights assigned to the first bone pose and the second bone pose respectively:
  • the radial basis function is used to interpolate between the first bone pose and the second bone pose to obtain the first bone pose and the second bone pose, respectively The weight assigned.
  • a radial basis function (RBF) can be applied to interpolate between the stored first bone pose and the second bone pose Through processing, the weights assigned to the first bone posture and the second bone posture respectively are obtained.
  • the above-mentioned weight based on the first bone pose and the weight of the second bone pose can be realized in this way, and the reference surface model and the example surface model are fused to obtain the corrected surface model of the target bone pose.
  • the modified surface model of the target bone pose can be obtained by means of weighted summation.
  • the weight assigned to P 0 is W 0
  • the weight assigned to P 1 is W 1 .
  • step 540 a virtual object is generated based on the target bone skin model and the modified surface model.
  • the above-mentioned skin model based on the target bone and the modified surface model can be realized in this way to generate a virtual object: determine the pose transformation matrix between the first bone pose and the target bone pose as the target pose Transformation matrix; based on the target pose transformation matrix, the modified surface model in the coordinate space is subjected to coordinate transformation processing to obtain the extracted modified surface model; the extracted modified surface model and the target bone skin model are superimposed to generate a virtual Object.
  • the stored reference surface model and the example surface model are located in the same coordinate space, and the corrected surface model is obtained by fusing the reference surface model and the example surface model. Therefore, the corrected surface model is also located in the coordinate space.
  • the modified surface model needs to be extracted from the coordinate space to conform to the coordinate space corresponding to the target bone posture.
  • the posture transformation matrix between the first bone posture and the target bone posture is determined.
  • the posture transformation matrix determined here is named the target posture transformation matrix.
  • the modified surface model in the coordinate space is subjected to coordinate transformation processing to obtain the extracted modified surface model, and then the extracted modified surface model and the target bone skin model are superimposed to generate Virtual object.
  • the extracted modified surface model can be located in the coordinate space corresponding to the target bone posture, which facilitates the generation of virtual objects.
  • the method before superimposing the extracted modified surface model and the target bone skin model, the method further includes: performing surface deformation processing on the target bone skin model.
  • the above-mentioned superimposed processing of the extracted modified surface model and the target bone skin model can be realized to generate a virtual object: the extracted modified surface model and the target bone skin model that has undergone surface deformation processing are processed Overlay processing to generate virtual objects.
  • the reference surface model is skinned based on the target bone posture.
  • the target bone skin model can be subjected to surface deformation processing according to actual needs, such as applying various complex Non-linear additional surface deformation.
  • the surface deformation can be generated based on a third-party system. For example, after the target skeletal skin model is generated, a third-party muscle simulation system is called to apply a specific shape of muscle on the target skeletal skin model.
  • the extracted corrected surface model After extracting the corrected surface model obtained by the fusion process from the coordinate space where the reference surface model is located, the extracted corrected surface model can be combined with the target bone skin model with additional surface deformation (that is, surface deformation processed) Perform overlay processing to generate virtual objects.
  • additional surface deformation that is, surface deformation processed
  • the method further includes: performing rendering processing on the generated virtual object to display the virtual object in the virtual scene.
  • the terminal device may perform rendering processing on the generated virtual object locally to display the virtual object in the virtual scene; when the electronic device used to generate the virtual object is a server At this time, the server may send the generated rendering data of the virtual object to the terminal device, and the terminal device may perform rendering processing according to the received rendering data to display the virtual object in the virtual scene.
  • the virtual scene can also be obtained through rendering.
  • FIG. 3C is a schematic flowchart of an image processing method provided by an embodiment of the present application.
  • Step 420 shown in FIG. 3B can be updated to step 610.
  • step 610 at least Two second bone poses.
  • only one example surface model can be generated, so as to generate virtual objects based on the reference surface model and one example surface model; at least two example surface models can also be generated according The reference surface model and at least two example surface models are used to generate virtual objects.
  • an example is given for the latter case.
  • at least two second bone poses of the virtual object are acquired.
  • the case where the acquired first bone posture is P 0 , the reference surface model is M 0 , and the acquired second bone posture includes P 1 , P 2, and P 3 will be described as an example.
  • step 430 shown in FIG. 3B can be updated to step 620.
  • step 620 at least two example surface models of the second bone poses are generated based on the reference surface model and the at least two second bone poses.
  • an example surface model of the second bone pose is generated based on the reference surface model and the second bone pose.
  • a corresponding example surface model can be obtained. For example, to obtain a second bone surface model posture Example 1 P M 1, a second bone surface model posture Example 2 P M 2, and a second bone surface model pose Example 3 P M 3.
  • step 440 shown in FIG. 3B can be updated to step 630.
  • step 630 the pose transformation matrix between the first bone pose and each second bone pose is determined.
  • a posture transformation matrix between the first skeleton posture and the second skeleton posture is also determined.
  • step 450 shown in FIG. 3B can be updated to step 640.
  • step 640 based on the pose transformation matrix between the first bone pose and each second bone pose, at least two example surface models and The reference surface model is converted to the same coordinate space.
  • the example surface model M 2 of the second bone pose P 2 can be converted to a reference
  • the surface model is the coordinate space where M 0 is located, and so on.
  • step 530 shown in FIG. 3B can be updated to step 650.
  • step 650 based on the reference surface model in the coordinate space and at least two example surface models, a modified surface model of the target bone pose is determined.
  • the generated material resources include at least two sample surface models.
  • the electronic device determines the modified surface model according to all stored sample surface models and reference surface models.
  • the weight W 0 assigned to P 0 can be determined .
  • the weight assigned to P1 is W 1
  • the weight assigned to P 2 is W 2
  • the weight assigned to P 3 is W 3 .
  • fusion processing is performed on M 0 , M 1 , M 2, and M 3 to obtain a corrected surface model M.
  • At least two second bone poses may be used to generate material resources and virtual objects. As the number of second bone poses increases, the effect of the generated virtual objects will be better.
  • the pre-generated material resources include the reference surface model M 0 , the sample surface model M 1 , the sample surface model M 2 , the sample surface model M 3 , and the sample surface model M i , where i is an integer greater than 3.
  • a first gesture includes skeletal bone attitude reference surface of the model M 0 P 0, the second example of bone surface model pose 1 M P 1, a second example of bone surface model posture M 2 P 2, the surface of sample second bone posture model M 3 P 3, the surface of the model sample up to a second bone posture M i P i.
  • the pose transformation matrix between the target bone pose P and the first bone pose P 0 is calculated, so as to perform coordinate space transformation in the subsequent process through the pose transformation matrix.
  • the reference surface model M 0 is skinned to obtain the target bone skin model. Furthermore, on the basis of the target bone skin model, a third-party system can be used to perform surface deformation processing, that is, additional surface deformation is applied, for example, a third-party muscle simulation system is used to apply a specific shape of muscle.
  • the RBF algorithm is used to perform interpolation processing between the read bone poses, so as to obtain the weights that need to be assigned to the read bone poses when all the read bone poses are merged into the target bone pose P. Furthermore, according to the assigned weights, the surface models corresponding to the read bone postures are fused to obtain the corrected surface model stored in the reference coordinate space.
  • Fig. 6 shows a terminal interface diagram in which the image processing method of the embodiment of the present application is applied to Maya software (a kind of three-dimensional modeling software) to generate virtual objects.
  • the pre-generated material resources may be upper arm surface models of various skeletal postures generated for the upper arm of the virtual object.
  • the terminal interface diagram of Maya software shown in the embodiment of the application includes four nodes integrated with corresponding functions: upper arm posture reading node (uperarm_r node), skin node (skinCluster node), and additional surface deformation node (deformate node) , Fusion node (posedriver node).
  • the upperarm_r node is mainly used to read the upper arm surface model of various bone poses in the material resources;
  • the shinCluster node is mainly used for skinning processing;
  • the deformate node is mainly used to apply additional surface deformation, that is, surface deformation processing;
  • posedriver node It is mainly used for fusion processing of surface models to generate modified surface models.
  • the calculation and application of the pose transformation matrix can be encapsulated in the posedriver node.
  • the skinCluster node used for skinning can only be placed in the virtual object generation phase.
  • the surface model is fused behind the posedriver node.
  • the deformate node used to apply additional surface deformation cannot be directly placed between the skinCluster node and the posedriver node.
  • the modified surface model obtained by the posedriver node can be compatible with any non-linear additional surface deformation, so that it can be directly superimposed on the skin after the coordinate conversion process is performed through the pose transformation matrix. On the bone skin model treated with additional surface deformation.
  • an embodiment of the present application also provides an image processing device, including: a first acquisition module 710, configured to acquire a reference surface model of a first bone pose of a virtual object; and a second acquisition module 720, configured to Obtain the second bone pose of the virtual object; the generation module 730 is configured to generate an example surface model of the second bone pose based on the reference surface model and the second bone pose; the matrix determination module 740 is configured to determine the first bone pose and the second bone pose The posture transformation matrix between the skeletal postures; the coordinate space conversion module 750 is configured to convert the example surface model and the reference surface model to the same coordinate space based on the posture transformation matrix, so as to perform processing based on the reference surface model and the example surface model in the coordinate space Generation of virtual objects.
  • the generation module 730 is configured to: perform skin processing on the reference surface model based on the second bone pose to obtain the example bone skin model of the second bone pose; extract the surface data of the example bone skin model according to the surface The data generates an example surface model of the second bone pose.
  • the image processing device is configured to: obtain the requester's modification request for the example skeleton skin model; when the modification request is verified, the modification authority of the example skeleton skin model is opened to the requester, so that The requester modifies the example skeleton skin model; extracts the surface data of the modified example skeleton skin model.
  • the coordinate space conversion module 750 is configured to perform coordinate conversion processing on the example surface model according to the pose transformation matrix to convert the example surface model to the coordinate space where the reference surface model is located.
  • the image processing device is configured to: obtain the target bone pose of the virtual object; perform skin processing on the reference surface model based on the target bone pose to obtain the target bone skin model of the target bone pose; and based on the reference in the coordinate space
  • the surface model and the sample surface model are used to determine the modified surface model of the target bone posture; based on the target bone skin model and the modified surface model, virtual objects are generated.
  • the image processing device is configured to: determine the weights assigned to the first bone pose and the second bone pose when the first bone pose and the second bone pose are merged into the target bone pose; based on the first bone The weight of the posture and the weight of the second bone posture are fused to the reference surface model and the example surface model to obtain a modified surface model of the target bone posture.
  • the image processing device is configured to perform interpolation processing between the first bone pose and the second bone pose by using a radial basis function when fusing the first bone pose and the second bone pose into the target bone pose , Get the weights assigned to the first bone posture and the second bone posture respectively.
  • the image processing device is configured to: use the weight of the first bone pose as the weight of the reference surface model, use the weight of the second bone pose as the weight of the example surface model, and perform processing on the reference surface model and the example surface model. After weighting, the modified surface model of the target bone pose is obtained.
  • the image processing device is configured to: determine the posture transformation matrix between the first bone posture and the target bone posture as the target posture transformation matrix; perform the correction surface model in the coordinate space based on the target posture transformation matrix Coordinate transformation is processed to obtain the extracted modified surface model; the extracted modified surface model and the target bone skin model are superimposed to generate a virtual object.
  • the image processing device is configured to: perform surface deformation processing on the target bone skin model; and perform superposition processing on the extracted corrected surface model and the target bone skin model subjected to surface deformation processing to generate a virtual object.
  • the number of second bone poses includes at least two; the image processing device is configured to generate at least two example surface models of the second bone poses based on the reference surface model and the at least two second bone poses; Determine the pose transformation matrix between the first bone pose and each second bone pose; convert at least two example surface models and reference surface models based on the pose transformation matrix between the first bone pose and each second bone pose To the same coordinate space.
  • the image processing device is configured to: obtain the target bone pose of the virtual object; perform skin processing on the reference surface model based on the target bone pose to obtain the target bone skin model of the target bone pose; and based on the reference in the coordinate space
  • the surface model and at least two sample surface models are used to determine the modified surface model of the target bone posture; based on the target bone skin model and the modified surface model, a virtual object is generated.
  • the image processing device is configured to: obtain manipulation data for the virtual object; and determine the target bone posture of the virtual object based on the manipulation data.
  • the image processing device is configured to: obtain sensor data; determine a user action based on the sensor data; and determine a target bone posture of the virtual object based on the user action.
  • the image processing apparatus is configured to perform rendering processing on the generated virtual object to display the virtual object in the virtual scene.
  • the electronic device 80 according to an embodiment of the present application will be described below with reference to FIG. 8.
  • the electronic device 80 shown in FIG. 8 is only an example, and should not bring any limitation to the function and scope of use of the embodiments of the present application.
  • the electronic device 80 is represented in the form of a general-purpose computing device.
  • the components of the electronic device 80 may include, but are not limited to: the aforementioned at least one processing unit 810, the aforementioned at least one storage unit 820, and a bus 830 connecting different system components (including the storage unit 820 and the processing unit 810).
  • the storage unit stores program codes, and the program codes can be executed by the processing unit 810, so that the processing unit 810 executes the steps according to various exemplary embodiments of the present application described in the description section of the foregoing exemplary method of this specification.
  • the processing unit 810 may perform various steps as shown in FIG. 3A, FIG. 3B, and FIG. 3C.
  • the storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 8201 and/or a cache storage unit 8202, and may further include a read-only storage unit (ROM) 8203.
  • RAM random access storage unit
  • ROM read-only storage unit
  • the storage unit 820 may also include a program/utility tool 8204 having a set of (at least one) program module 8205.
  • program module 8205 includes but is not limited to: an operating system, one or more application programs, other program modules, and program data, Each of these examples or some combination may include the implementation of a network environment.
  • the bus 830 may represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local area using any bus structure among multiple bus structures. bus.
  • the electronic device 80 can also communicate with one or more external devices 900 (such as keyboards, pointing devices, Bluetooth devices, etc.), and can also communicate with one or more devices that enable users to interact with the electronic device 80, and/or communicate with Any device (such as a router, modem, etc.) that enables the electronic device 80 to communicate with one or more other computing devices.
  • This communication can be performed through an input/output (I/O) interface 850.
  • An input/output (I/O) interface 850 is connected to the display unit 840.
  • the electronic device 80 may also communicate with one or more networks (for example, a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) through the network adapter 860.
  • networks for example, a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet
  • the network adapter 860 communicates with other modules of the electronic device 80 through the bus 830. It should be understood that although not shown in the figure, other hardware and/or software modules can be used in conjunction with the electronic device 80, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives And data backup storage system, etc.
  • the example embodiments described here can be implemented by software, or can be implemented by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, U disk, mobile hard disk, etc.) or on the network , Including several instructions to make a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) execute the method according to the embodiment of the present application.
  • a computing device which can be a personal computer, a server, a terminal device, or a network device, etc.
  • a computer-readable storage medium which stores computer-readable instructions.
  • the image processing method provided in the embodiments of the present application is implemented.
  • the embodiment of the present application also provides a program product for implementing the image processing method, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can run on a terminal device, such as a personal computer.
  • a program product of the present application is not limited thereto.
  • the readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, device, or device.
  • the program product can adopt any combination of one or more readable media.
  • the readable medium may be a readable signal medium or a readable storage medium.
  • the readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or a combination of any of the above. More specific examples (non-exhaustive list) of readable storage media include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RGM), read-only memory (ROM), erasable Type programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.
  • the computer-readable signal medium may include a data signal propagated in baseband or as a part of a carrier wave, and readable program code is carried therein. This propagated data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • the readable signal medium may also be any readable medium other than a readable storage medium, and the readable medium may send, propagate, or transmit a program for use by or in combination with the instruction execution system, apparatus, or device.
  • the program code contained on the readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.
  • the program code used to perform the operations of this application can be written in any combination of one or more programming languages.
  • Programming languages include object-oriented programming languages—such as JAVA, C++, etc., as well as conventional procedural programming. Language-such as "C" language or similar programming language.
  • the program code can be executed entirely on the user's computing device, partly on the user's device, executed as an independent software package, partly on the user's computing device and partly executed on the remote computing device, or entirely on the remote computing device or server Executed on.
  • the remote computing device can be connected to a user computing device through any kind of network, including a local area network (KGN) or a wide area network (WGN), or it can be connected to an external computing device (for example, using Internet service providers). Business to connect via the Internet).
  • KGN local area network
  • WGN wide area network
  • modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory.
  • the features and functions of two or more modules or units described above may be embodied in one module or unit.
  • the features and functions of a module or unit described above can be further divided into multiple modules or units to be embodied.
  • the example embodiments described here can be implemented by software, or can be implemented by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, U disk, mobile hard disk, etc.) or on the network , Including several instructions to make a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) execute the method according to the embodiment of the present application.
  • a non-volatile storage medium which can be a CD-ROM, U disk, mobile hard disk, etc.
  • Including several instructions to make a computing device which can be a personal computer, a server, a mobile terminal, or a network device, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Software Systems (AREA)
  • Computer Graphics (AREA)
  • Computer Hardware Design (AREA)
  • Human Computer Interaction (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Architecture (AREA)
  • Health & Medical Sciences (AREA)
  • Social Psychology (AREA)
  • Psychiatry (AREA)
  • Geometry (AREA)
  • General Health & Medical Sciences (AREA)
  • Processing Or Creating Images (AREA)

Abstract

一种图像处理方法、装置、电子设备及存储介质，所述方法包括：获取虚拟对象的第一骨骼姿态的基准表面模型(410)；获取虚拟对象的第二骨骼姿态(420)；基于基准表面模型及所述第二骨骼姿态，生成第二骨骼姿态的范例表面模型(430)；确定第一骨骼姿态与第二骨骼姿态之间的姿态变换矩阵(440)；基于所述姿态变换矩阵，将范例表面模型及所述基准表面模型转换至同一坐标空间，以根据坐标空间中的基准表面模型及范例表面模型进行虚拟对象的生成(450)。

Description

图像处理方法、装置、电子设备及存储介质
相关申请的交叉引用
本申请基于申请号为202010036516.7、申请日为2020年01月14日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请涉及图像处理领域，具体涉及一种图像处理方法、装置、电子设备及计算机可读存储介质。
背景技术
随着信息技术的快速发展，扩展了感知环境以及获取信息的渠道，尤其是虚拟化显示技术，能够根据实际需求实现虚拟对象的显示，具有各种典型的应用场景，例如在游戏虚拟场景中显示虚拟对象，以与用户进行交互。
在技术不断发展、更新的过程中，用户对虚拟对象的要求也越来越高，例如要求虚拟对象更加逼真、生动、具有艺术表现力。对于主要依赖于视觉输出信息的虚拟对象来说，为了满足用户要求，最直接的方式便是在视觉层面进行优化，因此，在虚拟对象的生成阶段，存在为虚拟对象施加各种表面形变(如让虚拟对象的某部分进行抖动)的需求。
发明内容
本申请实施例提供了一种图像处理方法，包括：
获取虚拟对象的第一骨骼姿态的基准表面模型；
获取所述虚拟对象的第二骨骼姿态；
基于所述基准表面模型及所述第二骨骼姿态，生成所述第二骨骼姿态的范例表面模型；
确定所述第一骨骼姿态与所述第二骨骼姿态之间的姿态变换矩阵；
基于所述姿态变换矩阵，将所述范例表面模型及所述基准表面模型转换至同一坐标空间，以根据所述坐标空间中的所述基准表面模型及所述范例表面模型进行所述虚拟对象的生成。
本申请实施例提供了一种图像处理装置，包括：
第一获取模块，配置为获取虚拟对象的第一骨骼姿态的基准表面模型；
第二获取模块，配置为获取所述虚拟对象的第二骨骼姿态；
生成模块，配置为基于所述基准表面模型及所述第二骨骼姿态，生成所述第二骨骼姿态的范例表面模型；
矩阵确定模块，配置为确定所述第一骨骼姿态与所述第二骨骼姿态之间的姿态变换矩阵；
坐标空间转换模块，配置为基于所述姿态变换矩阵，将所述范例表面模型及所述基准表面模型转换至同一坐标空间，以根据所述坐标空间中的所述基准表面模型及所述范例表面模型进行所述虚拟对象的生成。
本申请实施例提供了一种电子设备，包括：存储器，配置为存储有计算机可读指令；处理器，配置为执行所述存储器存储的所述计算机可读指令，以实现本申请实施例提供的图像处理方法。
本申请实施例提供了一种计算机可读存储介质，存储有计算机可读指令，当所述计算机可读指令被处理器执行时，实现本申请实施例提供的图像处理方法。
附图说明
通过参考附图详细描述其示例实施例，本申请的上述和其它目标、特征及优点将变得更加显而易见。
图1示出了本申请实施例提供的图像处理系统的架构示意图。
图2示出了本申请实施例提供的应用于游戏场景的图像处理系统的架构示意图。
图3A示出了本申请实施例提供的图像处理方法的流程图。
图3B示出了本申请实施例提供的图像处理方法的流程图。
图3C示出了本申请实施例提供的图像处理方法的流程图。
图4示出了本申请实施例提供的素材资源生成阶段的流程图。
图5示出了本申请实施例提供的虚拟对象生成阶段的流程图。
图6示出了本申请实施例提供的在Maya软件中进行虚拟对象生成的终端界面图。
图7示出了本申请实施例提供的图像处理装置的框图。
图8示出了本申请实施例提供的电子设备的硬件图。
具体实施方式
现在将参考附图更全面地描述示例实施方式。然而，示例实施方式能够以多种形式实施，且不应被理解为限于在此阐述的范例；相反，提供这些示例实施方式使得本申请的描述将更加全面和完整，并将示例实施方式的构思全面地传达给本领域的技术人员。附图仅为本申请的示意性图解，并非一定是按比例绘制。图中相同的附图标记表示相同或类似的部分，因而将省略对它们的重复描述。
此外，所描述的特征、结构或特性可以以任何合适的方式结合在一个或更多示例实施方式中。在下面的描述中，提供许多具体细节从而给出对本申请的示例实施方式的充分理解。然而，本领域技术人员将意识到，可以实践本申请的技术方案而省略特定细节中的一个或更多，或者可以采用其它的方法、组元、步骤等。在其它情况下，不详细示出或描述公知结构、方法、实现或者操作以避免喧宾夺主而使得本申请的各方面变得模糊。
附图中所示的一些方框图是功能实体，不一定必须与物理或逻辑上独立的实体相对应。可以采用软件形式来实现这些功能实体，或在一个或多个硬件模块或集成电路中实现这些功能实体，或在不同网络和/或处理器装置和/或微控制器装置中实现这些功能实体。
在以下的描述中，涉及到“一些实施例”，其描述了所有可能实施例的子集，但是可以理解,“一些实施例”可以是所有可能实施例的相同子集或不同子集，并且可以在不冲突的情况下相互结合。在以下的描述中，所涉及的术语“多个”是指至少两个。
除非另有定义，本文所使用的所有的技术和科学术语与属于本申请的技术领域的技术人员通常理解的含义相同。本文中所使用的术语只是为了描述本申请实施例的目的，不是旨在限制本申请。
首先对本申请实施例涉及到的部分概念进行简要说明。
1)虚拟场景：利用设备输出的区别于现实世界的场景，通过裸眼或设备的辅助能够形成对虚拟场景的视觉感知，例如通过显示屏幕输出的二维影像，通过立体投影、虚拟现实和增强现实技术等立体显示技术来输出的三维影像；此外，还可以通过各种可能的硬件形成听觉感知、触觉感知、嗅觉感知和运动感知等各种模拟现实世界的感知。
2)虚拟对象：虚拟场景中可以进行交互的各种人和物的形象，或在虚拟场景中的可活动对象，可以通过游戏引擎、绘图软件或者其他图像处理工具生成，用于生成虚拟对象的核心技术是三维图像技术。该可活动对象可以是虚拟人物、虚拟动物、动漫人物等，比如，在虚拟场景中显示的人物、动物、植物、油桶、墙壁、石块等。该虚拟对象可以是该虚拟场景中的一个虚拟的用于代表用户的虚拟形象。虚拟对象在虚拟场景中具有自身的形状和体积，占据虚拟场景中的一部分空间。在本申请实施例中，虚拟对象可以是动画角色，本申请实施例所提供的图像处理方法，可以适用于骨骼动画中动画角色的图像处理过程。
3)骨骼动画(Skeletal Animation)：是模型动画的一种。在骨骼动画中，虚拟对象(如动画角色)至少包含两种数据：骨骼姿态、表面模型。其中，骨骼姿态描述了虚拟对象“骨骼”的朝向和位置；表面模型描述了虚拟对象表面上的多个顶点，即表面模型可以理解为虚拟对象表面上的一系列顶点的集合。骨骼姿态的变化能够相应带动表面模型的变化，从而可以通过 驱动骨骼姿态，生成相应的动画。
4)蒙皮处理：为了实现“骨骼姿态的变化能够相应带动表面模型的变化”，需要定义特定骨骼能够驱动虚拟对象表面上的哪些顶点，而定义特定骨骼能够驱动虚拟对象表面上的哪些顶点(即是将特定的骨骼与虚拟对象表面上特定的顶点进行绑定)的过程，即称为蒙皮处理。
5)骨骼蒙皮模型：指根据骨骼姿态对表面模型进行蒙皮处理后，带有骨骼蒙皮信息的模型。骨骼蒙皮模型主要包括两部分信息：表面信息(表面数据)、骨骼蒙皮信息。
6)表面形变：在进行虚拟对象的生成时，可能会需要在虚拟对象上添加额外的表面形变，以满足特定的业务需求。例如，通过蒙皮处理以及其他的必要处理过程后，已经能够生成一个体型富态的动画角色，但在实际的业务需求中，要求该动画角色运动时，其腰上的赘肉能够像真实生活情况中一样进行抖动。甚至更进一步地，比真实生活情况更为夸张地进行抖动，以满足艺术需求。在这种情况下，就需要在动画角色上添加能够模拟赘肉抖动的额外表面形变。可以理解的，在实际应用中，额外表面形变可以是非线性的表面形变，甚至可以是难以捕捉形变规律的、复杂的非线性的表面形变。
7)素材资源：在本申请实施例中，在实际进行虚拟对象的生成之前，会预先进行相关素材资源的制作(生产)，素材资源包括基准表面模型及范例表面模型。其中，基准表面模型是预设的最基本的素材资源，可以只有一个；范例表面模型是在基准表面模型基础上进一步得到的素材资源，可以只有一个，也可以有至少两个。
随着技术发展，对虚拟对象的视觉表现的要求越来越高，在虚拟对象的生成阶段，存在为虚拟对象施加各种表面形变(如复杂的非线性表面形变)的需求。然而，在目前的骨骼动画技术中，无法支持在虚拟对象的生成阶段进行表面形变处理，对表面形变的兼容性低，无法满足上述需求。
针对于此，本申请实施例提供了一种图像处理方法、装置、电子设备及计算机可读存储介质，能够使生产素材资源的过程跳过表面形变的影响，从而支持在虚拟对象的生成阶段进行表面形变处理，提升对表面形变的兼容性。
作为示例，参见图1，图1示出了本申请实施例提供的图像处理系统的一个架构示意图，在图1中，图像处理系统包括第一终端设备10、服务器20及第二终端设备30。
在一些实施例中，以电子设备是终端设备为例，本申请实施例提供的图像处理方法可以由终端设备实现。为了便于理解，以第一终端设备10作为示例进行说明。例如，第一终端设备10可以执行素材资源的生成操作，其中，第一终端设备10可以向第一用户提供制作素材资源的相关功能，例 如确定表面模型的功能、调整骨骼姿态的功能等，以在第一用户的操作下生成素材资源。第一终端设备10可以将生成的素材资源存储在本地，并在需要时，根据存储的素材资源进行虚拟对象的生成，并对生成的虚拟对象进行渲染，以显示出虚拟对象，如在输出的虚拟场景中显示虚拟对象。该种方式适用于终端设备的计算能力较强的场景。
在一些实施例中，以电子设备是服务器为例，本申请实施例提供的图像处理方法可以由服务器实现。例如，服务器20可以进行素材资源的自动生成，例如根据预先存储的第一骨骼姿态、第二骨骼姿态以及基准表面模型生成素材资源，并将生成的素材资源存储在本地。服务器20还可以根据存储的素材资源进行虚拟对象的生成，并将生成的虚拟对象的渲染数据发送至终端设备，以使终端设备显示虚拟对象，如在终端设备输出的虚拟场景中显示虚拟对象。
在一些实施例中，本申请实施例提供的图像处理方法还可以由终端设备和服务器协同实现。例如，在素材资源的生成阶段，主要涉及到第一终端设备10和服务器20。其中，第一终端设备10主要作为前端，用于响应第一用户的操作，生成素材资源；服务器20主要作为后端，用于在素材资源的生成过程中为第一终端设备10提供相应的功能支持(如确定表面模型的功能、调整骨骼姿态的功能等)，并将第一终端设备10发送的素材资源进行存储。
在虚拟对象的生成阶段，主要涉及到第二终端设备30和服务器20。其中，服务器20主要作为后端，用于根据存储的素材资源进行虚拟对象的生成，并将生成的虚拟对象的渲染数据发送给第二终端设备30；第二终端设备30主要作为前端，用于根据从服务器20接收到的渲染数据进行渲染，以显示该虚拟对象，如在输出的虚拟场景中进行显示，便于第二用户观看或进行交互。作为示例，在图1中示出了第二终端设备30输出的虚拟场景31以及虚拟场景31中的虚拟对象32，虚拟场景31及虚拟对象32可以在第二终端设备30的图形界面中进行显示。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
作为示例，参见图2，图2示出了本申请实施例在游戏场景下的图像处理系统的一个架构示意图，图2中的图像处理系统包括模型师终端设备40、游戏服务器50及游戏玩家终端设备60。为了便于理解，以终端设备和服务器协同实现本申请实施例提供的图像处理方法的情况为例，进行说明。
在素材资源的生成阶段，主要涉及到模型师终端设备40、游戏服务器50。其中，模型师终端设备40主要作为前端，用于响应于模型师对游戏素材资源(例如自然站姿的游戏角色的表面模型、双臂张开姿态的游戏角色的表面模型等)的制作操作，生成游戏素材资源；游戏服务器50主要作为 后端，用于在游戏素材资源的生成过程中为模型师终端设备40提供相应的功能支持，并将模型师终端设备40发送的游戏素材资源进行存储。
其中，模型师可以在模型师终端设备40上进行相应的美术工作(例如通过数绘板修改游戏角色的表面模型，通过鼠标指令以及键盘指令修改游戏角色的表面模型等)，如此，模型师终端设备40可以在游戏服务器50的支持下，生成相应的游戏素材资源。
在虚拟对象的生成阶段，主要涉及到游戏玩家终端设备60及游戏服务器50。其中，游戏服务器50主要作为后端，用于根据存储的游戏素材资源进行虚拟对象的生成，这里以虚拟对象为游戏角色的情况作为示例，游戏服务器50还用于将生成的游戏角色的渲染数据发送给游戏玩家终端设备60；游戏玩家终端设备60主要作为前端，用于根据从游戏服务器50接收到的渲染数据进行渲染，以显示出该游戏角色，如在输出的游戏虚拟场景中进行显示。作为示例，在图2中示出了游戏玩家终端设备60输出的游戏虚拟场景61以及游戏虚拟场景61中的游戏角色62。
其中，游戏玩家可以于游戏玩家终端设备60进行游戏操作(例如通过鼠标控制游戏角色62的移动、通过键盘控制游戏角色62的技能释放、通过屏幕触摸控制游戏角色62的移动与技能释放等)，以对游戏玩家终端设备60显示的游戏角色62进行控制。在该过程中，游戏玩家终端设备60响应于玩家的游戏操作，产生相应的指令，并将指令发送至游戏服务器50。游戏服务器50可以根据接收到的指令，结合存储的游戏素材资源生成新的游戏角色，例如，图2中显示的游戏角色62是跳跃状态，生成的新的游戏角色可以是处于站立状态的游戏角色62，即骨骼姿态发生了改变。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
在一些实施例中，终端设备可以通过运行计算机程序来实现本申请实施例提供的图像处理方法，例如，计算机程序可以是操作系统中的原生程序或软件模块；可以是本地(Native)应用程序(APP，Application)，即需要在操作系统中安装才能运行的程序，例如游戏应用程序；也可以是小程序，即只需要下载到浏览器环境中就可以运行的程序；还可以是能够嵌入至任意APP中的小程序。总而言之，上述计算机程序可以是任意形式的应用程序、模块或插件。
本申请实施例可以借助于云技术(Cloud Technology)实现，云技术是指在广域网或局域网内将硬件、软件、网络等系列资源统一起来，实现数据的计算、储存、处理和共享的一种托管技术。在另一种意义上，云技术也是基于云计算商业模式应用的网络技术、信息技术、整合技术、管理平台技术、以及应用技术等的总称，可以组成资源池，按需所用，灵活便利。云计算技术将变成重要支撑。技术网络系统的后台服务需要大量的计算、存储资源。
在一些实施例中，上述的服务器可以是独立的物理服务器，也可以是多个物理服务器构成的服务器集群或者分布式系统，还可以是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、CDN、以及大数据和人工智能平台等基础云计算服务的云服务器，例如，云服务可以是生成素材资源的服务或生成虚拟对象的服务，以供终端设备调用。终端设备可以是智能手机、平板电脑、笔记本电脑、台式计算机、智能手表、智能电视等，但并不局限于此。终端设备以及服务器可以通过有线或无线通信方式进行直接或间接地连接，本申请实施例中不做限制。
将结合本申请实施例提供的电子设备的示例性应用和实施，说明本申请实施例提供的图像处理方法。
参见图3A，图3A是本申请实施例提供的图像处理方法的一个流程示意图，将结合图3A示出的步骤进行说明。
步骤410、获取虚拟对象的第一骨骼姿态的基准表面模型；
步骤420、获取虚拟对象的第二骨骼姿态；
步骤430、基于基准表面模型及第二骨骼姿态，生成第二骨骼姿态的范例表面模型；
步骤440、确定第一骨骼姿态与第二骨骼姿态之间的姿态变换矩阵；
步骤450、基于姿态变换矩阵，将范例表面模型及基准表面模型转换至同一坐标空间，以根据坐标空间中的基准表面模型及范例表面模型进行虚拟对象的生成。
本申请实施例中，在素材资源的生成阶段，电子设备在根据基准表面模型生成范例表面模型后，基于第一骨骼姿态(基准表面模型对应的骨骼姿态)与第二骨骼姿态(范例表面模型对应的骨骼姿态)之间的姿态变换矩阵，将范例表面模型与基准表面模型转换至同一坐标空间，得到同一坐标空间的素材资源。如此，使得素材资源的生成过程能够跳过额外表面形变的影响，从而在后续虚拟对象的生成阶段能够兼容额外的表面形变(如各种复杂的非线性的额外表面形变)，能够有效避免在不同坐标空间中进行处理所导致的结果不可控(如最终生成的虚拟对象与施加的表面形变并不相符)，提高了骨骼动画对表面形变的兼容性。
下面对素材资源的生成阶段的具体实施过程进行描述。
在步骤410中，获取虚拟对象的第一骨骼姿态的基准表面模型。
这里，虚拟对象是指虚拟化的各种人和物的对象，例如虚拟对象可以是动画角色，也可以是树木、旗帜或纸张等。在本申请实施例中，虚拟对象具有多种骨骼姿态，例如，动画角色具有自然站立姿态下的骨骼姿态及双臂水平姿态下的骨骼姿态等，旗帜具有静止时的骨骼姿态及被风吹拂时 的骨骼姿态等。
对于虚拟对象，获取虚拟对象的第一骨骼姿态的表面模型，以作为基准表面模型。例如，基准表面模型可以是虚拟对象在自然站姿下的表面模型，第一骨骼姿态即为虚拟对象在自然站姿下的骨骼姿态。其中，第一骨骼姿态可以仅有一个。
本申请实施例中，基准表面模型可以预先生成并在电子设备中进行存储，当需要进行素材资源的生成时，电子设备即可调取出该基准表面模型；基准表面模型也可以由用户(例如进行美术工作的模型师)实时于电子设备(如模型师终端设备)上制作，第一骨骼姿态同理。
在步骤420中，获取虚拟对象的第二骨骼姿态。
第二骨骼姿态指的是待生成的范例表面模型的骨骼姿态；范例表面模型指的是以基准表面模型为参考生成的、非第一骨骼姿态的表面模型。例如，基准表面模型为虚拟对象在自然站姿下的表面模型，则以虚拟对象在自然站姿下的表面模型为参考，生成的虚拟对象在双臂水平姿态下的表面模型即为一个范例表面模型。相应的，虚拟对象在双臂水平姿态下的骨骼姿态即为第二骨骼姿态。
本申请实施例中，第二骨骼姿态可以预先生成并在电子设备中进行存储；也可以由用户(例如进行美术工作的模型师)实时于电子设备上制作。本申请实施例对第二骨骼姿态的数量不做限定，例如可以是一个或至少两个，为了便于理解，图3A中以第二骨骼姿态仅包括一个的情况进行示例说明。
在步骤430中，基于基准表面模型及第二骨骼姿态，生成第二骨骼姿态的范例表面模型。
基于获取的基准表面模型及第二骨骼姿态，可以生成第二骨骼姿态的表面模型，为了便于区分，将这里生成的表面模型命名为范例表面模型。在一些实施例中，可以通过这样的方式来实现上述的基于基准表面模型及第二骨骼姿态，生成第二骨骼姿态的范例表面模型：基于第二骨骼姿态对基准表面模型进行蒙皮处理，得到第二骨骼姿态的范例骨骼蒙皮模型；提取范例骨骼蒙皮模型的表面数据，根据表面数据生成第二骨骼姿态的范例表面模型。
本申请实施例中，基于获取到的第二骨骼姿态，对获取到的基准表面模型进行蒙皮处理，得到第二骨骼姿态的骨骼蒙皮模型。其中，骨骼蒙皮模型主要包括两部分信息：表面数据、骨骼蒙皮信息。为了便于区分，将这里得到的骨骼蒙皮模型命名为范例骨骼蒙皮模型。
例如，基准表面模型为水平姿态的水平手臂模型，第二骨骼姿态为向上姿态的向上手臂的骨骼姿态。基于向上手臂的骨骼姿态对水平手臂模型进行蒙皮处理，得到带有骨骼蒙皮信息的、向上姿态的向上手臂模型，即范例骨骼蒙皮模型。
得到第二骨骼姿态的范例骨骼蒙皮模型后，提取出其表面数据，即可根据表面数据生成第二骨骼姿态的范例表面模型，其中，范例表面模型包括提取出的表面数据所描述的、虚拟对象表面上的各个顶点。通过上述的蒙皮处理的方式，能够自动得到第二骨骼姿态的范例表面模型，减少人力成本。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
在一些实施例中，提取范例骨骼蒙皮模型的表面数据之前，还包括：
获取请求方对范例骨骼蒙皮模型的修改请求；当对修改请求的验证通过时，将范例骨骼蒙皮模型的修改权限开放给请求方，以使请求方对范例骨骼蒙皮模型进行修改。
可以通过这样的方式实现上述的提取范例骨骼蒙皮模型的表面数据：提取修改后的范例骨骼蒙皮模型的表面数据。
在本申请实施例中，在对基准表面模型进行蒙皮处理得到范例骨骼蒙皮模型后，且在提取范例骨骼蒙皮模型中的表面数据得到范例表面模型之前，可以允许请求方(例如位于前端的模型师终端设备)对该范例骨骼蒙皮模型进行修改。
蒙皮处理本身是一个线性的数学处理过程，故而经由蒙皮处理得到的范例骨骼蒙皮模型只是做出了简单的骨骼姿态改变，范例骨骼蒙皮模型中的表面数据很可能无法满足实际需求，如特定的美术需求。例如，基于手肘弯曲的骨骼姿态对水平手臂模型(表面模型)进行蒙皮处理，能够得到带有骨骼蒙皮信息的、手肘弯曲姿态的弯曲手臂模型(骨骼蒙皮模型)。水平手臂模型中的肱二头肌的表面较为平坦，经过蒙皮处理后得到的弯曲手臂模型中肱二头肌的表面几乎不会发生明显变化，仍旧较为平坦。可以理解的，在实际生活中，手肘弯曲后，会对肌肉造成积压，相比水平手臂模型，弯曲手臂模型中肱二头肌的表面应当明显隆起。若有特定的美术需求，还会需要弯曲手臂模型中肱二头肌的表面的隆起程度更为明显。
因此，蒙皮处理后，可以支持请求方对范例蒙皮骨骼模型进行修改，以满足相应的需求，如美术需求。例如，用户通过请求方向电子设备(如服务器)发送针对范例骨骼蒙皮模型的修改请求，电子设备在对接收到的修改请求验证通过时，将相应的修改权限开放给请求方，使得用户能够于请求方对范例骨骼蒙皮模型进行修改。
以游戏场景举例，模型师根据艺术需要，通过模型师终端设备向游戏服务器发送修改请求，以请求对经过蒙皮处理得到的范例骨骼蒙皮模型进行修改。游戏服务器在对该修改请求验证通过时，将相应的修改权限开放给模型师终端设备，如此，模型师便可在模型师终端设备上，对范例骨骼蒙皮模型进行修改。
用户于请求方对范例骨骼蒙皮模型进行修改的过程中，可以修改范例 骨骼蒙皮模型中的表面数据；也可以修改范例骨骼蒙皮模型对应的第二骨骼姿态，来更新范例骨骼蒙皮模型中的表面数据。
修改完毕之后，用户可通过请求方，向电子设备发送修改完毕的消息，电子设备在接收到修改完毕的消息时，提取修改后的范例骨骼蒙皮模型中的表面数据，以根据该表面数据生成第二骨骼姿态的范例表面模型。通过上述的结合人为修改的方式，能够提升范例骨骼蒙皮模型中的表面数据的准确性，满足实际应用场景中的各种需求。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
在步骤440中，确定第一骨骼姿态与该第二骨骼姿态之间的姿态变换矩阵。
姿态变换矩阵指的是用于描述第一骨骼姿态与第二骨骼姿态之间的相互变换关系的数学矩阵。通过姿态变换矩阵，可以将第一骨骼姿态变换为第二骨骼姿态，也可以将第二骨骼姿态变换为第一骨骼姿态。
本申请实施例中，第一骨骼姿态与第二骨骼姿态在电子设备中均是以具体数据的形式存在的。例如，电子设备中可以存储骨骼顶点的三维空间坐标，通过调整骨骼顶点的三维空间坐标，即可得到对应的骨骼姿态，即骨骼姿态可以通过三维空间坐标数据的形式进行存储、处理。其中，一个骨骼姿态通常包括多个骨骼顶点，通过多个骨骼顶点，可以表示骨骼姿态中各个骨骼的朝向和位置。因此，可以将第一骨骼姿态表示为一个数学矩阵A，将第二骨骼姿态表示为一个数学矩阵B，根据A*C＝B的公式，可以计算得到矩阵C，即姿态变换矩阵。
在步骤450中，基于姿态变换矩阵，将范例表面模型与基准表面模型转换至同一坐标空间，以根据坐标空间中的基准表面模型与范例表面模型进行虚拟对象的生成。
本申请实施例中，将生成的素材资源转换至同一坐标空间中，以便进行虚拟对象的生成。即，在素材资源的生成阶段，将范例表面模型及基准表面模型转换至同一坐标空间。例如，可以将范例表面模型转换至基准表面模型所在的坐标空间，也可以将基准表面模型转换至范例表面模型所在的坐标空间。
在一些实施例中，可以通过这样的方式来实现上述的基于姿态变换矩阵，将范例表面模型及基准表面模型转换至同一坐标空间：根据姿态变换矩阵，对范例表面模型进行坐标变换处理，以将范例表面模型转换至基准表面模型所在的坐标空间。
在本申请实施例中，基准表面模型是人为设定的标准，因此，可以根据姿态变换矩阵，对范例表面模型进行坐标变换处理，以将范例表面模型转换至基准表面模型所处的坐标空间。进而，可以在基准表面模型所处的坐标空间中，存储基准表面模型及范例表面模型，完成素材资源的生成。
下面参考图4，对本申请实施例中素材资源的生成阶段的流程进行说明。
如图4所示，基准表面模型M 0位于一个坐标空间中，为了便于区分，将该坐标空间命名为基准坐标空间。电子设备初始获取到的数据包括基准表面模型M 0、基准表面模型M 0对应的第一骨骼姿态P 0、待生成的范例表面模型M i对应的第二骨骼姿态P i。
根据第一骨骼姿态P 0与第二骨骼姿态P i，计算这二者之间的姿态变换矩阵，以通过该姿态变换矩阵在后续过程中进行坐标空间的变换。
根据第二骨骼姿态P i对基准表面模型M 0进行蒙皮处理，得到范例骨骼蒙皮模型，进而可以由用户(如模型师)根据实际需要(如艺术需要)对该范例骨骼蒙皮模型进行修改，得到满足要求的新范例骨骼蒙皮模型(即修改后的范例骨骼蒙皮模型)。提取新范例骨骼蒙皮模型中的表面数据，从而生成第二骨骼姿态P i的范例表面模型，再应用姿态变换矩阵，把该范例表面模型变换到基准坐标空间，从而得到位于基准坐标空间中的范例表面模型M i。
本申请实施例中，电子设备通过姿态变换矩阵，将基准表面模型M 0及范例表面模型M i均存储在基准坐标空间，使得在后续的虚拟对象生成过程中，对于基准表面模型M 0及范例表面模型M i进行的一系列处理均在该基准坐标空间中进行，从而跳过了表面形变的影响。如此，使得生成的素材资源，即基准表面模型M 0与范例表面模型M i能够兼容任意的额外表面形变，如各种复杂的非线性的额外表面形变，能够有效避免在不同坐标空间中进行处理所导致的结果不可控，例如最终生成的虚拟对象与施加的表面形变并不相符。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
在一些实施例中，参见图3B，图3B是本申请实施例提供的图像处理方法的一个流程示意图，基于图3A，在步骤450之后，还可以在步骤510中，获取虚拟对象的目标骨骼姿态。
这里，目标骨骼姿态指的是所需要生成的虚拟对象的骨骼姿态，可以根据实际应用场景进行设定。例如，目标骨骼姿态为在双手叉腰站姿下的骨骼姿态，则最终生成的虚拟对象即为双手叉腰站姿。在本申请实施例中，目标骨骼姿态可以预先生成并在电子设备中进行存储，也可以实时获取。
在一些实施例中，可以通过这样的方式来实现上述的获取虚拟对象的目标骨骼姿态：获取针对虚拟对象的操控数据；基于操控数据，确定虚拟对象的目标骨骼姿态。
本申请实施例中，可以通过针对虚拟对象的操控数据，如基于特定的输入设备输入的操控数据，来确定目标骨骼姿态。例如，操控数据可以来 源于图1的第二终端设备30中的输入设备，或者图2的游戏玩家终端设备60中的输入设备。其中，操控数据可以包括键盘数据、鼠标数据及屏幕触摸数据中的至少一种，当然并不限于此。
以游戏场景进行举例说明，玩家可以在手机终端上进行屏幕触摸操作，从而指示手机终端运行的游戏中的游戏角色以特定姿态做出相应的动作。手机终端在检测到屏幕触摸操作时，将屏幕触摸操作触发的屏幕触摸数据上传至游戏服务器，游戏服务器基于获取到的屏幕触摸数据，确定玩家所指示的特定姿态，即，游戏角色的目标骨骼姿态。通过上述方式，能够使用户以主动操控的方式来确定目标骨骼姿态，从而控制虚拟对象的生成。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
在一些实施例中，可以通过这样的方式来实现上述的获取虚拟对象的目标骨骼姿态：获取传感器数据；基于传感器数据，确定用户动作；基于用户动作，确定虚拟对象的目标骨骼姿态。
本申请实施例中，还可以通过传感器数据来确定虚拟对象的目标骨骼姿态。其中，传感器数据可以来源于传感器设备，例如，图1的第二终端设备30中的传感器设备，或者图2的游戏玩家终端设备60中的传感器设备。传感器数据可以包括有关定位信息的定位传感器数据、以及有关局部肢体动作的动作传感器数据中的至少一种。
以游戏场景进行举例说明，玩家佩戴与电脑终端相通信的虚拟现实(Virtual Reality，VR)设备，例如VR眼镜或VR手柄等，通过做出特定动作，从而指示电脑终端中运行的VR游戏中的相应游戏角色做出同样的动作。VR设备对应的VR传感器系统(例如由至少两个基站组成的传感器系统)可以通过监控VR设备中的传感器(例如VR手柄中的传感器)，生成相应的传感器数据，并将传感器数据发送至电脑终端。电脑终端将传感器数据上传至服务器后，服务器即可根据传感器数据确定用户动作(即实现动作捕捉)，并将确定出的用户动作所对应的骨骼姿态，作为游戏角色的目标骨骼姿态。通过上述方式，使用户通过自身做出动作的方式，来确定目标骨骼姿态，能够给用户带来更大的沉浸感，提升用户的交互体验。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。另外，上述的确定虚拟对象的目标骨骼姿态的方式，同样适用于确定虚拟对象的第一骨骼姿态及第二骨骼姿态。
在步骤520中，基于目标骨骼姿态对基准表面模型进行蒙皮处理，得到目标骨骼姿态的目标骨骼蒙皮模型。
在得到目标骨骼姿态后，基于目标骨骼姿态对基准表面模型进行蒙皮处理，得到目标骨骼姿态的骨骼蒙皮模型，为了便于区分，将这里得到的骨骼蒙皮模型命名为目标骨骼蒙皮模型。
在步骤530中，基于坐标空间中的基准表面模型及范例表面模型，确 定目标骨骼姿态的修正表面模型。
这里，基于同一坐标空间中的基准表面模型及范例表面模型，确定目标骨骼姿态的修正表面模型。修正表面模型指的是能够对目标骨骼蒙皮模型中的表面数据做出一定程度的修正，以在一定程度上满足实际需求(如美术需求)的表面模型。修正表面模型对应的骨骼姿态与目标骨骼蒙皮模型对应的骨骼姿态基本一致，即为目标骨骼姿态。
在一些实施例中，基于预先生成的属于特定美术风格的素材资源，生成的目标骨骼姿态的表面模型(即修正表面模型)，同样属于该美术风格，如此，能够保证基于不同目标骨骼姿态生成的虚拟对象的风格一致，提升统一性。
在一些实施例中，可以通过这样的方式来实现上述的基于坐标空间中的基准表面模型及范例表面模型，确定目标骨骼姿态的修正表面模型：确定在将第一骨骼姿态及第二骨骼姿态融合为目标骨骼姿态时，分别为第一骨骼姿态及第二骨骼姿态分配的权重；基于第一骨骼姿态的权重及第二骨骼姿态的权重，对基准表面模型及范例表面模型进行融合处理，得到目标骨骼姿态的修正表面模型。
这里，确定分别为第一骨骼姿态及第二骨骼姿态分配的权重，其中，根据分别为第一骨骼姿态及第二骨骼姿态分配的权重，能够将第一骨骼姿态及第二骨骼姿态融合为目标骨骼姿态，这里的融合可以是加权求和，当然也可以是其他的方式。进而，基于第一骨骼姿态的权重及第二骨骼姿态的权重，对第一骨骼姿态的基准表面模型及第二骨骼姿态的范例表面模型进行融合处理，得到目标骨骼姿态的修正表面模型。
在一些实施例中，可以通过这样的方式来实现上述的确定在将第一骨骼姿态及第二骨骼姿态融合为目标骨骼姿态时，分别为第一骨骼姿态及第二骨骼姿态分配的权重：在将第一骨骼姿态及第二骨骼姿态融合为目标骨骼姿态时，通过径向基函数在第一骨骼姿态与第二骨骼姿态之间进行插值处理，得到分别为第一骨骼姿态及第二骨骼姿态分配的权重。
这里，在将第一骨骼姿态及第二骨骼姿态融合为目标骨骼姿态时，可以应用径向基函数(Radial Basis Function，RBF)，在存储的第一骨骼姿态与第二骨骼姿态之间进行插值处理，从而得到分别为第一骨骼姿态及第二骨骼姿态分配的权重。
在一些实施例中，可以通过这样的方式来实现上述的基于第一骨骼姿态的权重及第二骨骼姿态的权重，对基准表面模型及范例表面模型进行融合处理，得到目标骨骼姿态的修正表面模型：将第一骨骼姿态的权重作为基准表面模型的权重，将第二骨骼姿态的权重作为范例表面模型的权重，并对基准表面模型及范例表面模型进行加权处理，得到目标骨骼姿态的修正表面模型。
这里，可以通过加权求和的方式，得到目标骨骼姿态的修正表面模型。 例如，在将第一骨骼姿态P 0及第二骨骼姿态P 1融合为目标骨骼姿态P时，为P 0分配的权重为W 0，为P 1分配的权重为W 1。进而，根据W 0及W 1，对第一骨骼姿态的基准表面模型M 0、以及第二骨骼姿态P 1的范例表面模型M 1进行融合处理，得到修正表面模型M，其中，M＝W 0*M 0+W 1*M 1。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制，除了加权求和的方式外，还可通过其他方式对基准表面模型及范例表面模型进行融合。
在步骤540中，基于目标骨骼蒙皮模型及修正表面模型，生成虚拟对象。
在一些实施例中，可以通过这样的方式来实现上述的基于目标骨骼蒙皮模型及修正表面模型，生成虚拟对象：确定第一骨骼姿态与目标骨骼姿态之间的姿态变换矩阵，以作为目标姿态变换矩阵；基于目标姿态变换矩阵，对坐标空间中的修正表面模型进行坐标变换处理，得到提取出的修正表面模型；将提取出的修正表面模型与目标骨骼蒙皮模型进行叠加处理，以生成虚拟对象。
本申请实施例中，存储的基准表面模型及范例表面模型位于同一坐标空间中，修正表面模型是通过基准表面模型与范例表面模型融合得到，因此，修正表面模型也位于该坐标空间中。在目标骨骼姿态与第一骨骼姿态不同的情况下，需要将修正表面模型从该坐标空间中提取出来，以符合目标骨骼姿态对应的坐标空间。例如，确定第一骨骼姿态与目标骨骼姿态之间的姿态变换矩阵，为了便于区分，将这里确定出的姿态变换矩阵命名为目标姿态变换矩阵。基于得到的目标姿态变换矩阵，对坐标空间中的修正表面模型进行坐标变换处理，得到提取出的修正表面模型，然后，将提取出的修正表面模型与目标骨骼蒙皮模型进行叠加处理，以生成虚拟对象。通过上述方式，使得提取出的修正表面模型能够位于目标骨骼姿态对应的坐标空间，便于进行虚拟对象的生成。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
在一些实施例中，将提取出的修正表面模型与目标骨骼蒙皮模型进行叠加处理之前，还包括：对目标骨骼蒙皮模型进行表面形变处理。可以通过这样的方式来实现上述的将提取出的修正表面模型与目标骨骼蒙皮模型进行叠加处理，以生成虚拟对象：将提取出的修正表面模型与经过表面形变处理的目标骨骼蒙皮模型进行叠加处理，以生成虚拟对象。
本申请实施例中，基于目标骨骼姿态对基准表面模型进行蒙皮处理，得到目标骨骼蒙皮模型后，可以根据实际需求，在目标骨骼蒙皮模型上进行表面形变处理，例如施加各种复杂的非线性的额外表面形变。其中，表面形变可以基于第三方系统生成，例如，生成目标骨骼蒙皮模型后，调用第三方的肌肉模拟系统，以在该目标骨骼蒙皮模型上施加特定形态的肌肉。
将融合处理得到的修正表面模型从基准表面模型所在的坐标空间中提取出来后，便可以将该提取出的修正表面模型与添加了额外表面形变(即经过表面形变处理)的目标骨骼蒙皮模型进行叠加处理，以生成虚拟对象。通过上述方式，能够满足各种各样的表面形变需求，提升对实际应用场景的适用性。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
在一些实施例中，步骤540之后，还包括：对生成的虚拟对象进行渲染处理，以在虚拟场景中显示虚拟对象。
这里，当用于生成虚拟对象的电子设备是终端设备时，终端设备可以在本地对生成的虚拟对象进行渲染处理，以在虚拟场景中显示虚拟对象；当用于生成虚拟对象的电子设备是服务器时，服务器可以将生成的虚拟对象的渲染数据发送至终端设备，终端设备可以根据接收到的渲染数据进行渲染处理，以在虚拟场景中显示虚拟对象。需要说明的是，虚拟场景也可以通过渲染得到。通过上述方式，提升了显示虚拟对象的灵活性。
如图3B所示，由于基准表面模型及范例表面模型均位于经由姿态变换矩阵构造出的同一独立的坐标空间中，因此，对基准表面模型及范例表面模型的处理均在该坐标空间内进行，不会受到表面形变的影响，从而使得在虚拟对象的生成阶段，能够兼容任意非线性的表面形变，提升对表面形变的兼容性。
在一些实施例中，参见图3C，图3C是本申请实施例提供的图像处理方法的一个流程示意图，图3B示出的步骤420可以更新为步骤610，在步骤610中，获取虚拟对象的至少两个第二骨骼姿态。
在本申请实施例中，在素材资源的生成阶段，可以仅生成一个范例表面模型，从而根据基准表面模型及一个范例表面模型进行虚拟对象的生成；也可以生成至少两个范例表面模型，从而根据基准表面模型及至少两个范例表面模型进行虚拟对象的生成。这里，针对后者情况进行举例说明，首先，获取虚拟对象的至少两个第二骨骼姿态。为了便于理解，以获取到的第一骨骼姿态为P 0，基准表面模型为M 0，获取到的第二骨骼姿态包括P 1、P 2及P 3的情况，进行举例说明。
在图3C中，图3B示出的步骤430可以更新为步骤620，在步骤620中，基于基准表面模型及至少两个第二骨骼姿态，分别生成至少两个第二骨骼姿态的范例表面模型。
针对获取到的每个第二骨骼姿态，基于基准表面模型及第二骨骼姿态，生成该第二骨骼姿态的范例表面模型。如此，对于每一个第二骨骼姿态，均可得到对应的范例表面模型。例如，可以得到第二骨骼姿态P 1的范例表面模型M 1、第二骨骼姿态P 2的范例表面模型M 2、以及第二骨骼姿态P 3的 范例表面模型M 3。
在图3C中，图3B示出的步骤440可以更新为步骤630，在步骤630中，确定第一骨骼姿态与每个第二骨骼姿态之间的姿态变换矩阵。
针对每个第二骨骼姿态，还确定第一骨骼姿态与第二骨骼姿态之间的姿态变换矩阵。
在图3C中，图3B示出的步骤450可以更新为步骤640，在步骤640中，基于第一骨骼姿态与每个第二骨骼姿态之间的姿态变换矩阵，将至少两个范例表面模型及基准表面模型转换至同一坐标空间。
这里，以第二骨骼姿态P 1为例，可以基于第一骨骼姿态P 0与第二骨骼姿态P 1之间的姿态变换矩阵，将第二骨骼姿态P 2的范例表面模型M 2转换至基准表面模型为M 0所在的坐标空间，以此类推。
在图3C中，图3B示出的步骤530可以更新为步骤650，在步骤650中，基于坐标空间中的基准表面模型及至少两个范例表面模型，确定目标骨骼姿态的修正表面模型。
本申请实施例中，生成的素材资源中包括至少两个范例表面模型。相应的，在根据生成的素材资源确定修正表面模型时，电子设备根据存储的所有范例表面模型及基准表面模型确定修正表面模型。
举例来说，可以确定将第一骨骼姿态P 0、第二骨骼姿态P 1、第二骨骼姿态P 2以及第二骨骼姿态P 3融合为目标骨骼姿态P时，为P 0分配的权重W 0，为P1分配的权重W 1，为P 2分配的权重W 2，为P 3分配的权重W 3。进而，根据W 0、W 1、W 2及W 3，对M 0、M 1、M 2及M 3进行融合处理，得到修正表面模型M。其中，可以采取W 0*M 0+W 1*M 1+W 2*M 2+W 3*M 3＝M的方式，来得到修正表面模型M。
如图3C所示，本申请实施例可以通过至少两个第二骨骼姿态来进行素材资源及虚拟对象的生成，随着第二骨骼姿态的数量增多，生成的虚拟对象的效果也会越好。
下面参考图5，对本申请实施例中虚拟对象的生成阶段的流程进行说明。
如图5所示，预先生成的素材资源(均存储于基准坐标空间)包括基准表面模型M 0、范例表面模型M 1、范例表面模型M 2、范例表面模型M 3、一直到范例表面模型M i，这里的i为大于3的整数。相应的，获取到的骨骼姿态包括基准表面模型M 0的第一骨骼姿态P 0、范例表面模型M 1的第二骨骼姿态P 1、范例表面模型M 2的第二骨骼姿态P 2、范例表面模型M 3的第二骨骼姿态P 3、一直到范例表面模型M i的第二骨骼姿态P i。
在获取到目标骨骼姿态P时，计算目标骨骼姿态P与第一骨骼姿态P 0这二者之间的姿态变换矩阵，以通过该姿态变换矩阵在后续过程中进行坐标空间的变换。
根据目标骨骼姿态P对基准表面模型M 0进行蒙皮处理，得到目标骨骼 蒙皮模型。进而，可以在目标骨骼蒙皮模型的基础上，应用第三方系统进行表面形变处理，即施加额外表面形变，例如应用第三方肌肉模拟系统施加特定形态肌肉。
读取存储的骨骼姿态，包括第一骨骼姿态P 0、第二骨骼姿态P 1、第二骨骼姿态P 2、第二骨骼姿态P 3、一直到第二骨骼姿态P i。应用RBF算法，在读取到的骨骼姿态间进行插值处理，从而得到将读取到的所有骨骼姿态融合为目标骨骼姿态P时，需要为读取到的各个骨骼姿态分配的权重。进而根据分配的权重，对读取到的各个骨骼姿态所对应的表面模型进行融合处理，得到存储于基准坐标空间的修正表面模型。
应用姿态变换矩阵，对存储于基准坐标空间的修正表面模型进行坐标变换处理，以将修正表面模型从基准坐标空间中提取出来，进而与经过蒙皮处理后得到的、并且施加了额外表面形变的目标骨骼蒙皮模型进行叠加处理，从而得到对应的虚拟对象。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
图6示出了将本申请实施例的图像处理方法应用于Maya软件(一种三维建模软件)，以进行虚拟对象生成的终端界面图。
在图6中，预先生成的素材资源可以是针对虚拟对象的上臂所生成的各种骨骼姿态的上臂表面模型。本申请实施例示出的Maya软件的终端界面图中，包括四个集成了相应功能的节点：上臂姿态读取节点(uperarm_r节点)、蒙皮节点(skinCluster节点)、额外表面形变节点(deformate节点)、融合节点(posedriver节点)。其中，uperarm_r节点主要用于读取素材资源中的各种骨骼姿态的上臂表面模型；shinCluster节点主要用于进行蒙皮处理；deformate节点主要用于施加额外表面形变，即进行表面形变处理；posedriver节点主要用于进行表面模型的融合处理，以生成修正表面模型。其中，posedriver节点中可以封装入姿态变换矩阵的计算与应用。
在相关技术中，由于生成的素材资源无法支持在虚拟对象的生成阶段兼容额外表面形变，因此，在虚拟对象的生成阶段中，用于进行蒙皮处理的skinCluster节点只能放在主要用于进行表面模型的融合处理的posedriver节点的后面。而且，在相关技术中，用于施加额外表面形变的deformate节点是无法直接放在skinCluster节点与posedriver节点之间的。根据本申请实施例所提供的图像处理方法，使得posedriver节点得到的修正表面模型能够兼容任意非线性的额外表面形变，从而在通过姿态变换矩阵进行坐标转换处理之后，能够直接叠加到已经过蒙皮处理、且施加了额外表面形变的骨骼蒙皮模型上。
需要说明的是，该实施例只是示例性的说明，不应对本申请的功能和使用范围造成限制。
如图7所示，本申请实施例还提供了一种图像处理装置，包括：第一获取模块710，配置为获取虚拟对象的第一骨骼姿态的基准表面模型；第二获取模块720，配置为获取虚拟对象的第二骨骼姿态；生成模块730，配置为基于基准表面模型及第二骨骼姿态，生成第二骨骼姿态的范例表面模型；矩阵确定模块740，配置为确定第一骨骼姿态与第二骨骼姿态之间的姿态变换矩阵；坐标空间转换模块750，配置为基于姿态变换矩阵，将范例表面模型及基准表面模型转换至同一坐标空间，以根据坐标空间中的基准表面模型与范例表面模型进行虚拟对象的生成。
在一些实施例中，生成模块730配置为：基于第二骨骼姿态对基准表面模型进行蒙皮处理，得到第二骨骼姿态的范例骨骼蒙皮模型；提取范例骨骼蒙皮模型的表面数据，根据表面数据生成第二骨骼姿态的范例表面模型。
在一些实施例中，图像处理装置配置为：获取请求方对范例骨骼蒙皮模型的修改请求；当对修改请求的验证通过时，将范例骨骼蒙皮模型的修改权限开放给请求方，以使请求方对范例骨骼蒙皮模型进行修改；提取修改后的范例骨骼蒙皮模型的表面数据。
在一些实施例中，坐标空间转换模块750配置为：根据姿态变换矩阵，对范例表面模型进行坐标变换处理，以将范例表面模型转换至基准表面模型所在的坐标空间。
在一些实施例中，图像处理装置配置为：获取虚拟对象的目标骨骼姿态；基于目标骨骼姿态对基准表面模型进行蒙皮处理，得到目标骨骼姿态的目标骨骼蒙皮模型；基于坐标空间中的基准表面模型及范例表面模型，确定目标骨骼姿态的修正表面模型；基于目标骨骼蒙皮模型及修正表面模型，生成虚拟对象。
在一些实施例中，图像处理装置配置为：确定在将第一骨骼姿态及第二骨骼姿态融合为目标骨骼姿态时，分别为第一骨骼姿态及第二骨骼姿态分配的权重；基于第一骨骼姿态的权重及第二骨骼姿态的权重，对基准表面模型及范例表面模型进行融合处理，得到目标骨骼姿态的修正表面模型。
在一些实施例中，图像处理装置配置为：在将第一骨骼姿态及第二骨骼姿态融合为目标骨骼姿态时，通过径向基函数在第一骨骼姿态与第二骨骼姿态之间进行插值处理，得到分别为第一骨骼姿态及第二骨骼姿态分配的权重。
在一些实施例中，图像处理装置配置为：将第一骨骼姿态的权重作为基准表面模型的权重，将第二骨骼姿态的权重作为范例表面模型的权重，并对基准表面模型及范例表面模型进行加权处理，得到目标骨骼姿态的修正表面模型。
在一些实施例中，图像处理装置配置为：确定第一骨骼姿态与目标骨 骼姿态之间的姿态变换矩阵，以作为目标姿态变换矩阵；基于目标姿态变换矩阵，对坐标空间中的修正表面模型进行坐标变换处理，得到提取出的修正表面模型；将提取出的修正表面模型与目标骨骼蒙皮模型进行叠加处理，以生成虚拟对象。
在一些实施例中，图像处理装置配置为：对目标骨骼蒙皮模型进行表面形变处理；将提取出的修正表面模型与经过表面形变处理的目标骨骼蒙皮模型进行叠加处理，以生成虚拟对象。
在一些实施例中，第二骨骼姿态的数量包括至少两个；图像处理装置配置为：基于基准表面模型及至少两个第二骨骼姿态，分别生成至少两个第二骨骼姿态的范例表面模型；确定第一骨骼姿态与每个第二骨骼姿态之间的姿态变换矩阵；基于第一骨骼姿态与每个第二骨骼姿态之间的姿态变换矩阵，将至少两个范例表面模型及基准表面模型转换至同一坐标空间。
在一些实施例中，图像处理装置配置为：获取虚拟对象的目标骨骼姿态；基于目标骨骼姿态对基准表面模型进行蒙皮处理，得到目标骨骼姿态的目标骨骼蒙皮模型；基于坐标空间中的基准表面模型及至少两个范例表面模型，确定目标骨骼姿态的修正表面模型；基于目标骨骼蒙皮模型及修正表面模型，生成虚拟对象。
在一些实施例中，图像处理装置配置为：获取针对虚拟对象的操控数据；基于操控数据，确定虚拟对象的目标骨骼姿态。
在一些实施例中，图像处理装置配置为：获取传感器数据；基于传感器数据，确定用户动作；基于用户动作，确定虚拟对象的目标骨骼姿态。
在一些实施例中，图像处理装置配置为：对生成的虚拟对象进行渲染处理，以在虚拟场景中显示虚拟对象。
下面参考图8来描述根据本申请实施例的电子设备80。图8显示的电子设备80仅仅是一个示例，不应对本申请实施例的功能和使用范围带来任何限制。
如图8所示，电子设备80以通用计算设备的形式表现。电子设备80的组件可以包括但不限于：上述至少一个处理单元810、上述至少一个存储单元820、连接不同系统组件(包括存储单元820和处理单元810)的总线830。
其中，存储单元存储有程序代码，程序代码可以被处理单元810执行，使得处理单元810执行本说明书上述示例性方法的描述部分中描述的根据本申请各种示例性实施方式的步骤。例如，处理单元810可以执行如图3A、图3B及图3C中所示的各个步骤。
存储单元820可以包括易失性存储单元形式的可读介质，例如随机存取存储单元(RAM)8201和/或高速缓存存储单元8202，还可以进一步包括只读存储单元(ROM)8203。
存储单元820还可以包括具有一组(至少一个)程序模块8205的程序/实用工具8204，这样的程序模块8205包括但不限于：操作系统、一个或者多个应用程序、其它程序模块以及程序数据，这些示例中的每一个或某种组合中可能包括网络环境的实现。
总线830可以为表示几类总线结构中的一种或多种，包括存储单元总线或者存储单元控制器、外围总线、图形加速端口、处理单元或者使用多种总线结构中的任意总线结构的局域总线。
电子设备80也可以与一个或多个外部设备900(例如键盘、指向设备、蓝牙设备等)通信，还可与一个或者多个使得用户能与该电子设备80交互的设备通信，和/或与使得该电子设备80能与一个或多个其它计算设备进行通信的任何设备(例如路由器、调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口850进行。输入/输出(I/O)接口850与显示单元840相连。并且，电子设备80还可以通过网络适配器860与一个或者多个网络(例如局域网(LAN)，广域网(WAN)和/或公共网络，例如因特网)通信。如图所示，网络适配器860通过总线830与电子设备80的其它模块通信。应当明白，尽管图中未示出，可以结合电子设备80使用其它硬件和/或软件模块，包括但不限于：微代码、设备驱动器、冗余处理单元、外部磁盘驱动阵列、RAID系统、磁带驱动器以及数据备份存储系统等。
通过以上的实施方式的描述，本领域的技术人员易于理解，这里描述的示例实施方式可以通过软件实现，也可以通过软件结合必要的硬件的方式来实现。因此，根据本申请实施方式的技术方案可以以软件产品的形式体现出来，该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM，U盘，移动硬盘等)中或网络上，包括若干指令以使得一台计算设备(可以是个人计算机、服务器、终端装置、或者网络设备等)执行根据本申请实施方式的方法。
在本申请的示例性实施例中，还提供了一种计算机可读存储介质，存储有计算机可读指令，当计算机可读指令被处理器执行时，实现本申请实施例提供的图像处理方法。
本申请实施例还提供了一种用于实现图像处理方法的程序产品，其可以采用便携式紧凑盘只读存储器(CD-ROM)并包括程序代码，并可以在终端设备，例如个人电脑上运行。然而，本申请的程序产品不限于此，在本申请实施例中，可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
程序产品可以采用一个或多个可读介质的任意组合。可读介质可以是可读信号介质或者可读存储介质。可读存储介质例如可以为但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。可读存储介质的更具体的例子(非穷举的列表)包括：具有一个或 多个导线的电连接、便携式盘、硬盘、随机存取存储器(RGM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。
计算机可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了可读程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。可读信号介质还可以是可读存储介质以外的任何可读介质，该可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。
可读介质上包含的程序代码可以用任何适当的介质传输，包括但不限于无线、有线、光缆、RF等等，或者上述的任意合适的组合。
可以以一种或多种程序设计语言的任意组合来编写用于执行本申请操作的程序代码，程序设计语言包括面向对象的程序设计语言—诸如JAVA、C++等，还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算设备上执行、部分地在用户设备上执行、作为一个独立的软件包执行、部分在用户计算设备上部分在远程计算设备上执行、或者完全在远程计算设备或服务器上执行。在涉及远程计算设备的情形中，远程计算设备可以通过任意种类的网络，包括局域网(KGN)或广域网(WGN)，连接到用户计算设备，或者，可以连接到外部计算设备(例如利用因特网服务提供商来通过因特网连接)。
应当注意，尽管在上文详细描述中提及了用于动作执行的设备的若干模块或者单元，但是这种划分并非强制性的。实际上，根据本申请的实施方式，上文描述的两个或更多模块或者单元的特征和功能可以在一个模块或者单元中具体化。反之，上文描述的一个模块或者单元的特征和功能可以进一步划分为由多个模块或者单元来具体化。
此外，尽管在附图中以特定顺序描述了本申请中方法的各个步骤，但是，这并非要求或者暗示必须按照该特定顺序来执行这些步骤，或是必须执行全部所示的步骤才能实现期望的结果。附加的或备选的，可以省略某些步骤，将多个步骤合并为一个步骤执行，以及/或者将一个步骤分解为多个步骤执行等。
通过以上的实施方式的描述，本领域的技术人员易于理解，这里描述的示例实施方式可以通过软件实现，也可以通过软件结合必要的硬件的方式来实现。因此，根据本申请实施方式的技术方案可以以软件产品的形式体现出来，该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM，U盘，移动硬盘等)中或网络上，包括若干指令以使得一台计算设备(可以是个人计算机、服务器、移动终端、或者网络设备等)执行根据本申请实施方式的方法。
本领域技术人员在考虑说明书及实践这里公开的申请后，将容易想到本申请的其它实施方案。本申请旨在涵盖本申请的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本申请的一般性原理并包括本申请未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的，本申请的真正范围和精神由所附的权利要求指出。

Claims (19)

  1. 一种图像处理方法，由电子设备执行，所述方法包括：
    获取虚拟对象的第一骨骼姿态的基准表面模型；
    获取所述虚拟对象的第二骨骼姿态；
    基于所述基准表面模型及所述第二骨骼姿态，生成所述第二骨骼姿态的范例表面模型；
    确定所述第一骨骼姿态与所述第二骨骼姿态之间的姿态变换矩阵；
    基于所述姿态变换矩阵，将所述范例表面模型及所述基准表面模型转换至同一坐标空间，以根据所述坐标空间中的所述基准表面模型及所述范例表面模型进行所述虚拟对象的生成。
  2. 根据权利要求1所述的方法，其中，所述基于所述基准表面模型及所述第二骨骼姿态，生成所述第二骨骼姿态的范例表面模型，包括：
    基于所述第二骨骼姿态对所述基准表面模型进行蒙皮处理，得到所述第二骨骼姿态的范例骨骼蒙皮模型；
    提取所述范例骨骼蒙皮模型的表面数据，根据所述表面数据生成所述第二骨骼姿态的范例表面模型。
  3. 根据权利要求2所述的方法，其中，所述提取所述范例骨骼蒙皮模型的表面数据之前，所述方法还包括：
    获取请求方对所述范例骨骼蒙皮模型的修改请求；
    当对所述修改请求的验证通过时，将所述范例骨骼蒙皮模型的修改权限开放给所述请求方，以使所述请求方对所述范例骨骼蒙皮模型进行修改；
    所述提取所述范例骨骼蒙皮模型的表面数据，包括：
    提取修改后的所述范例骨骼蒙皮模型的表面数据。
  4. 根据权利要求1所述的方法，其中，所述基于所述姿态变换矩阵，将所述范例表面模型及所述基准表面模型转换至同一坐标空间，包括：
    根据所述姿态变换矩阵，对所述范例表面模型进行坐标变换处理，以将所述范例表面模型转换至所述基准表面模型所在的坐标空间。
  5. 根据权利要求1所述的方法，其中，所述方法还包括：
    获取所述虚拟对象的目标骨骼姿态；
    基于所述目标骨骼姿态对所述基准表面模型进行蒙皮处理，得到所述目标骨骼姿态的目标骨骼蒙皮模型；
    基于所述坐标空间中的所述基准表面模型及所述范例表面模型，确定所述目标骨骼姿态的修正表面模型；
    基于所述目标骨骼蒙皮模型及所述修正表面模型，生成所述虚拟对象。
  6. 根据权利要求5所述的方法，其中，所述基于所述坐标空间中的所述基准表面模型及所述范例表面模型，确定所述目标骨骼姿态的修正表面模型，包括：
    确定在将所述第一骨骼姿态及所述第二骨骼姿态融合为所述目标骨骼姿态时，分别为所述第一骨骼姿态及所述第二骨骼姿态分配的权重；
    基于所述第一骨骼姿态的权重及所述第二骨骼姿态的权重，对所述基准表面模型及所述范例表面模型进行融合处理，得到所述目标骨骼姿态的修正表面模型。
  7. 根据权利要求6所述的方法，其中，所述确定在将所述第一骨骼姿态及所述第二骨骼姿态融合为所述目标骨骼姿态时，分别为所述第一骨骼姿态及所述第二骨骼姿态分配的权重，包括：
    在将所述第一骨骼姿态及所述第二骨骼姿态融合为所述目标骨骼姿态时，通过径向基函数在所述第一骨骼姿态与所述第二骨骼姿态之间进行插值处理，得到分别为所述第一骨骼姿态及所述第二骨骼姿态分配的权重。
  8. 根据权利要求6所述的方法，其中，所述基于所述第一骨骼姿态的权重及所述第二骨骼姿态的权重，对所述基准表面模型及所述范例表面模型进行融合处理，得到所述目标骨骼姿态的修正表面模型，包括：
    将所述第一骨骼姿态的权重作为所述基准表面模型的权重，将所述第二骨骼姿态的权重作为所述范例表面模型的权重，并对所述基准表面模型及所述范例表面模型进行加权处理，得到所述目标骨骼姿态的修正表面模型。
  9. 根据权利要求5所述的方法，其中，所述基于所述目标骨骼蒙皮模型及所述修正表面模型，生成所述虚拟对象，包括：
    确定所述第一骨骼姿态与所述目标骨骼姿态之间的姿态变换矩阵，以作为目标姿态变换矩阵；
    基于所述目标姿态变换矩阵，对所述坐标空间中的所述修正表面模型进行坐标变换处理，得到提取出的所述修正表面模型；
    将提取出的所述修正表面模型与所述目标骨骼蒙皮模型进行叠加处理，以生成所述虚拟对象。
  10. 根据权利要求9所述的方法，其中，所述将提取出的所述修正表面模型与所述目标骨骼蒙皮模型进行叠加处理之前，所述方法还包括：
    对所述目标骨骼蒙皮模型进行表面形变处理；
    所述将提取出的所述修正表面模型与所述目标骨骼蒙皮模型进行叠加处理，以生成所述虚拟对象，包括：
    将提取出的所述修正表面模型与经过表面形变处理的所述目标骨骼蒙皮模型进行叠加处理，以生成所述虚拟对象。
  11. 根据权利要求5所述的方法，其中，所述获取所述虚拟对象的目标骨骼姿态，包括：
    获取针对所述虚拟对象的操控数据；
    基于所述操控数据，确定所述虚拟对象的目标骨骼姿态。
  12. 根据权利要求5所述的方法，其中，所述获取所述虚拟对象的目标骨骼姿态，包括：
    获取传感器数据；
    基于所述传感器数据，确定用户动作；
    基于所述用户动作，确定所述虚拟对象的目标骨骼姿态。
  13. 根据权利要求1至12任一项所述的方法，其中，
    所述第二骨骼姿态的数量包括至少两个；
    所述基于所述基准表面模型及所述第二骨骼姿态，生成所述第二骨骼姿态的范例表面模型，包括：
    基于所述基准表面模型及至少两个所述第二骨骼姿态，分别生成至少两个所述第二骨骼姿态的范例表面模型。
  14. 根据权利要求13所述的方法，其中，
    所述确定所述第一骨骼姿态与所述第二骨骼姿态之间的姿态变换矩阵，包括：
    确定所述第一骨骼姿态与每个所述第二骨骼姿态之间的姿态变换矩阵；
    所述基于所述姿态变换矩阵，将所述范例表面模型及所述基准表面模型转换至同一坐标空间，包括：
    基于所述第一骨骼姿态与每个所述第二骨骼姿态之间的姿态变换矩阵，将至少两个所述范例表面模型及所述基准表面模型转换至同一坐标空间。
  15. 根据权利要求14所述的方法，其中，所述方法还包括：
    获取所述虚拟对象的目标骨骼姿态；
    基于所述目标骨骼姿态对所述基准表面模型进行蒙皮处理，得到所述目标骨骼姿态的目标骨骼蒙皮模型；
    基于所述坐标空间中的所述基准表面模型及至少两个所述范例表面模型，确定所述目标骨骼姿态的修正表面模型；
    基于所述目标骨骼蒙皮模型及所述修正表面模型，生成所述虚拟对象。
  16. 根据权利要求1至12任一项所述的方法，其中，所述方法还包括：
    对生成的所述虚拟对象进行渲染处理，以在虚拟场景中显示所述虚拟对象。
  17. 一种图像处理装置，所述装置包括：
    第一获取模块，配置为获取虚拟对象的第一骨骼姿态的基准表面模型；
    第二获取模块，配置为获取所述虚拟对象的第二骨骼姿态；
    生成模块，配置为基于所述基准表面模型及所述第二骨骼姿态，生成所述第二骨骼姿态的范例表面模型；
    矩阵确定模块，配置为确定所述第一骨骼姿态与所述第二骨骼姿态之间的姿态变换矩阵；
    坐标空间转换模块，配置为基于所述姿态变换矩阵，将所述范例表面模型及所述基准表面模型转换至同一坐标空间，以根据所述坐标空间中的所述基准表面模型及所述范例表面模型进行所述虚拟对象的生成。
  18. 一种电子设备，包括：
    存储器，配置为存储有计算机可读指令；
    处理器，配置为执行所述存储器存储的所述计算机可读指令，以实现权利要求1至16中任一项所述的图像处理方法。
  19. 一种计算机可读存储介质，存储有计算机可读指令，当所述计算机可读指令被处理器执行时，实现权利要求1至16中任一项所述的图像处理方法。
PCT/CN2020/125081 2020-01-14 2020-10-30 图像处理方法、装置、电子设备及存储介质 Ceased WO2021143278A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2022520106A JP7299414B2 (ja) 2020-01-14 2020-10-30 画像処理方法、装置、電子機器及びコンピュータプログラム
EP20913250.5A EP4002286A4 (en) 2020-01-14 2020-10-30 Image processing method and apparatus, and electronic device and storage medium
KR1020227003382A KR102698917B1 (ko) 2020-01-14 2020-10-30 이미지 처리 방법 및 장치, 전자 장치 및 저장 매체
US17/706,242 US12183036B2 (en) 2020-01-14 2022-03-28 Image processing method and apparatus, electronic device, and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010036516.7A CN111223171B (zh) 2020-01-14 2020-01-14 图像处理方法、装置、电子设备及存储介质
CN202010036516.7 2020-01-14

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/706,242 Continuation US12183036B2 (en) 2020-01-14 2022-03-28 Image processing method and apparatus, electronic device, and storage medium

Publications (1)

Publication Number Publication Date
WO2021143278A1 true WO2021143278A1 (zh) 2021-07-22

Family

ID=70806354

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/125081 Ceased WO2021143278A1 (zh) 2020-01-14 2020-10-30 图像处理方法、装置、电子设备及存储介质

Country Status (6)

Country Link
US (1) US12183036B2 (zh)
EP (1) EP4002286A4 (zh)
JP (1) JP7299414B2 (zh)
KR (1) KR102698917B1 (zh)
CN (1) CN111223171B (zh)
WO (1) WO2021143278A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12183036B2 (en) 2020-01-14 2024-12-31 Tencent Technology (Shenzhen) Company Limited Image processing method and apparatus, electronic device, and storage medium

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111696184B (zh) * 2020-06-10 2023-08-29 上海米哈游天命科技有限公司 骨骼蒙皮融合确定方法、装置、设备和存储介质
CN112509098B (zh) * 2020-11-30 2024-02-13 北京百度网讯科技有限公司 动画形象生成方法、装置及电子设备
CN113034691B (zh) * 2021-03-22 2025-04-29 广州虎牙科技有限公司 人体模型的骨骼绑定方法、装置及电子设备
CN113101666B (zh) * 2021-05-07 2024-03-08 网易（杭州）网络有限公司 游戏角色模型方法、装置、计算机设备及存储介质
CN113610992B (zh) * 2021-08-04 2022-05-20 北京百度网讯科技有限公司 骨骼驱动系数确定方法、装置、电子设备及可读存储介质
CN115082604B (zh) 2022-07-06 2025-09-19 北京字跳网络技术有限公司 图像处理方法、装置、电子设备及存储介质
CN115526968A (zh) * 2022-10-14 2022-12-27 网易（杭州）网络有限公司 模型表情的重定向方法及装置、存储介质、电子设备
CN115937373B (zh) * 2022-12-23 2023-10-03 北京百度网讯科技有限公司 虚拟形象驱动方法、装置、设备以及存储介质
CN119971490B (zh) * 2023-11-10 2025-12-16 网易（杭州）网络有限公司 游戏的资源传输方法、装置、终端设备和存储介质
CN117456611B (zh) * 2023-12-22 2024-03-29 拓世科技集团有限公司 一种基于人工智能的虚拟人物训练方法及系统
CN118012272B (zh) * 2024-04-09 2024-06-07 南京维赛客网络科技有限公司 基于语境分析切换虚拟人物动作的方法、系统及存储介质
CN120339562B (zh) * 2025-06-23 2025-08-26 中汽研汽车检验中心（天津）有限公司 一种人体骨骼模型的姿态变换方法、装置、介质及设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130127853A1 (en) * 2011-11-17 2013-05-23 Mixamo, Inc. System and method for automatic rigging of three dimensional characters for facial animation
CN109395390A (zh) * 2018-10-26 2019-03-01 网易（杭州）网络有限公司 游戏角色脸部模型的处理方法、装置、处理器及终端
US20200005537A1 (en) * 2018-06-29 2020-01-02 Dreamworks Animation Llc Real-time collision deformation
CN111223171A (zh) * 2020-01-14 2020-06-02 腾讯科技（深圳）有限公司 图像处理方法、装置、电子设备及存储介质

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100530244C (zh) * 2005-06-21 2009-08-19 中国科学院计算技术研究所 一种基于骨骼的任意拓扑结构虚拟角色的驱动方法
US20070268293A1 (en) * 2006-05-19 2007-11-22 Erick Miller Musculo-skeletal shape skinning
JP5322078B2 (ja) 2007-03-26 2013-10-23 サミー株式会社 画像生成装置および画像生成プログラム
JP4579964B2 (ja) * 2007-12-13 2010-11-10 株式会社コナミデジタルエンタテインメント 画像生成装置、画像生成方法、ならびに、プログラム
US8400455B2 (en) * 2008-01-11 2013-03-19 Sony Corporation Method and apparatus for efficient offset curve deformation from skeletal animation
CN104021584B (zh) * 2014-06-25 2017-06-06 无锡梵天信息技术股份有限公司 一种骨骼蒙皮动画的实现方法
JP6340313B2 (ja) * 2014-12-24 2018-06-06 ソフトバンク株式会社 モデリングシステム、モデリングプログラム、及びモデリング方法
US9928663B2 (en) * 2015-07-27 2018-03-27 Technische Universiteit Delft Skeletal joint optimization for linear blend skinning deformations utilizing skeletal pose sampling
US10061871B2 (en) * 2015-07-27 2018-08-28 Technische Universiteit Delft Linear blend skinning weight optimization utilizing skeletal pose sampling
CN106228592A (zh) * 2016-09-12 2016-12-14 武汉布偶猫科技有限公司 一种服装三维模型自动绑定蒙皮信息的方法
WO2018095273A1 (zh) * 2016-11-24 2018-05-31 腾讯科技（深圳）有限公司 图像合成方法、装置以及匹配实现方法、装置
CN111640176A (zh) 2018-06-21 2020-09-08 华为技术有限公司 一种物体建模运动方法、装置与设备
WO2020009800A1 (en) * 2018-07-02 2020-01-09 Magic Leap, Inc. Methods and systems for interpolation of disparate inputs
CN110102050B (zh) * 2019-04-30 2022-02-18 腾讯科技（深圳）有限公司 虚拟对象显示方法、装置、电子设备及存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130127853A1 (en) * 2011-11-17 2013-05-23 Mixamo, Inc. System and method for automatic rigging of three dimensional characters for facial animation
US20200005537A1 (en) * 2018-06-29 2020-01-02 Dreamworks Animation Llc Real-time collision deformation
CN109395390A (zh) * 2018-10-26 2019-03-01 网易（杭州）网络有限公司 游戏角色脸部模型的处理方法、装置、处理器及终端
CN111223171A (zh) * 2020-01-14 2020-06-02 腾讯科技（深圳）有限公司 图像处理方法、装置、电子设备及存储介质

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
BLENDER CHINESE COMMUNITY: "Auto-Rig Pro Blender rig kit", 20 December 2019 (2019-12-20), pages 1 - 5, XP055828882, Retrieved from the Internet <URL:https://www.blendercn.org/6378.html> *
See also references of EP4002286A4

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12183036B2 (en) 2020-01-14 2024-12-31 Tencent Technology (Shenzhen) Company Limited Image processing method and apparatus, electronic device, and storage medium

Also Published As

Publication number Publication date
CN111223171A (zh) 2020-06-02
JP2022550555A (ja) 2022-12-02
US20220215583A1 (en) 2022-07-07
KR102698917B1 (ko) 2024-08-23
JP7299414B2 (ja) 2023-06-27
EP4002286A1 (en) 2022-05-25
KR20220028069A (ko) 2022-03-08
EP4002286A4 (en) 2022-11-30
CN111223171B (zh) 2025-06-27
US12183036B2 (en) 2024-12-31

Similar Documents

Publication Publication Date Title
WO2021143278A1 (zh) 图像处理方法、装置、电子设备及存储介质
US11823315B2 (en) Animation making method and apparatus, computing device, and storage medium
CN109377544B (zh) 一种人脸三维图像生成方法、装置和可读介质
CN113946211A (zh) 基于元宇宙的多个对象的交互方法及相关设备
JP2023549747A (ja) 触覚オブジェクトの表現フォーマット
US11928762B2 (en) Asynchronous multi-user real-time streaming of web-based image edits using generative adversarial network(s)
CN113223121A (zh) 视频生成方法、装置、电子设备及存储介质
CN109035415B (zh) 虚拟模型的处理方法、装置、设备和计算机可读存储介质
CN114332317B (zh) 动画数据处理方法、装置、程序产品、介质及电子设备
CN114399424A (zh) 模型训练方法及相关设备
CN114998490B (zh) 一种虚拟对象生成方法、装置、设备及存储介质
CN111447379B (zh) 生成信息的方法和装置
CN117826989A (zh) 电力元宇宙的拓展现实沉浸式交互方法及装置
CN116993868A (zh) 动画生成方法、装置、电子设备及存储介质
CN116431001A (zh) 一种在虚拟空间中实现ai交互的方法
CN115775300A (zh) 人体模型的重建方法、人体重建模型的训练方法及装置
JP2023011742A (ja) 画像処理方法及び装置、モデルトレーニング方法及び装置、電子機器、記憶媒体並びにコンピュータプログラム
HK40024180A (zh) 图像处理方法、装置、电子设备及存储介质
CN116246014B (zh) 一种形象生成方法、装置、存储介质及电子设备
CN120276589A (zh) 可实现现实虚拟的用户交互方法及装置
HK40024180B (zh) 图像处理方法、装置、电子设备及存储介质
CN121213841A (zh) 一种虚拟环境中衣物的动态更新方法、装置、设备及介质
CN116259099A (zh) 手势姿态估计方法、系统及计算机可读存储介质
HK40024182B (zh) 一种制作动画的方法、装置及存储介质
HK40024182A (zh) 一种制作动画的方法、装置及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20913250

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 20227003382

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2020913250

Country of ref document: EP

Effective date: 20220218

ENP Entry into the national phase

Ref document number: 2022520106

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE