WO2022078014A1 - 虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质 - Google Patents

虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质 Download PDF

Info

Publication number
WO2022078014A1
WO2022078014A1 PCT/CN2021/109551 CN2021109551W WO2022078014A1 WO 2022078014 A1 WO2022078014 A1 WO 2022078014A1 CN 2021109551 W CN2021109551 W CN 2021109551W WO 2022078014 A1 WO2022078014 A1 WO 2022078014A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
hand
target
virtual
wearing object
Prior art date
Application number
PCT/CN2021/109551
Other languages
English (en)
French (fr)
Inventor
卢艺帆
Original Assignee
北京字节跳动网络技术有限公司
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 北京字节跳动网络技术有限公司 filed Critical 北京字节跳动网络技术有限公司
Publication of WO2022078014A1 publication Critical patent/WO2022078014A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/107—Static hand or arm
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0641—Shopping interfaces
    • G06Q30/0643—Graphical representation of items or shoppers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements

Definitions

  • Embodiments of the present disclosure relate to the field of computer technologies, and in particular, to a virtual wearing object matching method, apparatus, electronic device, and computer-readable medium.
  • Some embodiments of the present disclosure propose a virtual wearing object matching method, apparatus, electronic device, and computer-readable medium to solve the technical problems mentioned in the above background section.
  • some embodiments of the present disclosure provide a virtual wearing object matching method, the method includes: acquiring an image to be processed and a wearing object image, the to-be-processed image includes a hand image, and the wearing object image includes wearing object parameters Recognize the above-mentioned image to be processed, and obtain the key point coordinates of the above-mentioned hand image; adjust the parameters of the initial hand model based on the above-mentioned key point coordinates, and obtain the target hand model corresponding to the above-mentioned hand image; Virtual wearing object; matching the virtual wearing object with the target hand model to generate a matching image of the virtual wearing object.
  • some embodiments of the present disclosure provide a virtual wearing object matching device, the device includes: an image acquisition unit configured to acquire a to-be-processed image and a wearing object image, the to-be-processed image includes a hand image, the above The wearing object image includes wearing object parameters; the key point coordinate acquisition unit is configured to identify the above-mentioned image to be processed, and obtain the key point coordinates of the above-mentioned hand image; the target hand model acquisition unit is configured to be based on the above-mentioned key point coordinates.
  • some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; and a memory on which one or more programs are stored, when the one or more programs are stored by the one or more programs described above When executed by the plurality of processors, the above-mentioned one or more processors are caused to execute the above-mentioned virtual wearing object matching method of the first aspect.
  • some embodiments of the present disclosure provide a computer-readable medium on which a computer program is stored, characterized in that, when the program is executed by a processor, the virtual wearing object matching method of the first aspect is implemented.
  • One of the above-mentioned embodiments of the present disclosure has the following beneficial effects: firstly, the image to be processed and the image of the wearing object are acquired, the coordinates of key points of the hand image in the image to be processed are determined, and the parameters of the hand are determined; The parameters of the initial hand model are adjusted by the coordinates of the key points to obtain the target hand model, which realizes the virtualization of the hand; after that, the virtual wearing object is constructed through the wearing object parameters, and the virtual wearing object is realized; finally, the virtual wearing object and the target The hand models are matched to generate matching images. The accuracy and effectiveness of the virtual matching of the hand and the wearing object are realized, and the effectiveness of the user selecting items through the network is improved.
  • FIG. 1 is a schematic diagram of an application scenario of a virtual wearing object matching method according to some embodiments of the present disclosure
  • FIG. 2 is a flowchart of some embodiments of a virtual wearable object matching method according to the present disclosure
  • FIG. 3 is a flowchart of other embodiments of the virtual wearing object matching method according to the present disclosure.
  • FIG. 4 is a flowchart of further embodiments of a virtual wearable object matching method according to the present disclosure
  • FIG. 5 is a schematic structural diagram of some embodiments of a virtual wearing object matching device according to the present disclosure.
  • FIG. 6 is a schematic structural diagram of an electronic device suitable for implementing some embodiments of the present disclosure.
  • FIG. 1 is a schematic diagram of an application scenario of a virtual wearing object matching method according to some embodiments of the present disclosure.
  • the electronic device 101 acquires the to-be-processed image including the hand image and the ring image (that is, the wearing object image), it first identifies the to-be-processed image to obtain The keypoint coordinates of the hand image in the image to be processed.
  • the key points may be multiple designated feature points of the hand (for example, the joints of fingers).
  • the coordinates of the key points may be the coordinates of each key point on the coordinate plane after the coordinate plane is constructed with the set point in the image to be processed as the origin (for example, it may be the sitting angle of the image to be processed).
  • the coordinates of the key points of the present disclosure can be used to represent the relative positional relationship between the set positions of the hands corresponding to the above hand images. For example, if there are 3 key points on a certain finger, the relative distance between the coordinates of the key points corresponding to the 3 key points (ie relative positional relationship) can represent information such as the length of the finger. Then, the electronic device 101 can adjust the parameters of the pre-built initial hand model according to the acquired coordinates of the key points, so that the parameters of the target hand model after the modified parameters are the same as the parameters of the actual hand corresponding to the hand image. Afterwards, the electronic device 101 may further construct a ring model by using the ring parameters, and render the ring model to obtain a virtual ring.
  • the electronic device 101 can match the virtual ring to the target hand model.
  • the electronic device 101 can generate a matching image. In this way, the virtual matching between the ring and the hand is realized, and the effectiveness of the user selecting items through the network is improved.
  • FIG. 1 It should be understood that the number of electronic devices in FIG. 1 is merely illustrative. There can be any number of terminal devices according to implementation needs.
  • the virtual wearing object matching method includes the following steps:
  • Step 201 acquiring the image to be processed and the image of the wearing object.
  • the execution body of the virtual wearing object matching method may acquire the image to be processed and the wearing object image through a wired connection or a wireless connection.
  • the wearing object image may be an image obtained by the user from the network, and is usually an image of an object that the user expects to wear.
  • the image to be processed may be an image acquired by a user through various image acquisition devices (for example, a smartphone).
  • the above wireless connection methods may include but are not limited to 3G/4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or developed in the future .
  • the to-be-processed image including the hand image and the wearing item image can be sent to the execution subject at the same time.
  • the wearing object image may include wearing object parameters.
  • the wearing object parameter may be information obtained by the user from the description of the wearing object on the network.
  • Step 202 Identify the above-mentioned image to be processed, and obtain the coordinates of key points of the above-mentioned hand image.
  • the execution subject can recognize the hand image from the image to be processed. Then, the execution body can use a certain point on the image to be processed as the origin to construct a plane coordinate system, and then the execution body can search for key points on the hand image and determine the coordinates of the key points in the plane coordinate system.
  • Step 203 Adjust parameters of the initial hand model based on the coordinates of the above-mentioned key points to obtain a target hand model corresponding to the above-mentioned hand image.
  • the executive body can determine the distance relationship between each key point of the hand in the hand image through the coordinates of the key point. Then, the execution subject can adjust the parameters of the initial hand model through the coordinates of the key points, so that the initial hand model is as close as possible to the actual hand size corresponding to the hand image, which is beneficial to improve the matching accuracy of the wearing object and the hand.
  • Step 204 constructing a virtual wearing object according to the above-mentioned wearing object parameters.
  • the executive body can also construct the wearing object model through the wearing object parameters. Then, the execution subject obtains the virtual wearable object by means of rendering or the like.
  • Step 205 Match the virtual wearing object with the target hand model to generate a matching image of the virtual wearing object.
  • the virtualized processing of the hand image and the wearing object is completed.
  • the virtual wearing object and the target hand model are the same as the wearing object and the hand in terms of size and shape, respectively.
  • the executive body can match the virtual wearing object with the above-mentioned target hand model.
  • the execution subject may generate a matching image of the virtual wearing object. In this way, the virtual matching between the wearing object and the hand is realized, and the effectiveness of the user selecting an object through the network is improved.
  • the image to be processed and the image of the wearing object are first obtained, and the coordinates of key points of the hand image in the image to be processed are determined, and the parameters of the hand are determined; Adjust the parameters of the initial hand model, obtain the target hand model, and realize the virtualization of the hand; then construct the virtual wearing object through the wearing object parameters, and realize the virtualization of the wearing object; finally, the virtual wearing object and the target hand model are analyzed. Matching is performed to generate matching images. The accuracy and effectiveness of the virtual matching of the hand and the wearing object are realized, and the effectiveness of the user selecting items through the network is improved.
  • the virtual wearing object matching method includes the following steps:
  • Step 301 acquiring the image to be processed and the image of the wearing object.
  • step 301 The content of step 301 is the same as that of step 201, and details are not repeated here.
  • Step 302 Identify the above-mentioned image to be processed, and obtain the coordinates of key points of the above-mentioned hand image.
  • step 302 The content of step 302 is the same as that of step 202, and details are not repeated here.
  • the above-mentioned recognizing the above-mentioned image to be processed, and obtaining the key point coordinates of the above-mentioned hand image may include the following steps:
  • the first step is to set at least one designated mark point corresponding to the above-mentioned set position on the above-mentioned hand image.
  • the execution subject can set at least one designated mark point on the hand image according to the needs of the initial hand model.
  • the set position can be each joint point of the hand.
  • the joint points can represent various dimensions of the hand, which is beneficial to improve the accuracy of the target hand model.
  • the coordinates of the key points of the hand image are determined based on the size information and the at least one designated marker point.
  • the image to be processed may also include size information.
  • the above-mentioned size information may be used to indicate the actual size of the hand corresponding to the above-mentioned hand image. That is, the size information may be information obtained by the user by measuring the actual size of the hand. For example, if the user wants to buy a ring, when sending the image to be processed to the execution subject, the size information of the user's hand can be sent together.
  • the size information may be: the length of the hand, the length of each finger, the width of the fingers, the thickness of the palm, and the like. Usually, the size information is more accurate, and the error when specifying the marker point is relatively large.
  • the execution body can determine the key point coordinates by combining the size information and the specified marker points. In this way, the accuracy of obtaining the coordinates of the key points is improved.
  • the above-mentioned recognizing the above-mentioned image to be processed and obtaining the key point coordinates of the above-mentioned hand image may include: searching for a target point cloud corresponding to the above-mentioned set position in the above-mentioned point cloud data data, and mark the coordinates of the above target point cloud data as key point coordinates.
  • the to-be-processed image acquired by the user is a three-dimensional image acquired by a point cloud data acquisition device
  • the to-be-processed image may include point cloud data.
  • the point cloud data is used to indicate the three-dimensional coordinates of the hand in space corresponding to the hand image.
  • the execution body can directly query the target point cloud data corresponding to the set position, and mark the coordinates of the target point cloud data as key point coordinates.
  • the coordinates of the key points at this time are three-dimensional coordinates, which can more conveniently adjust the parameters of the initial hand model.
  • Step 303 Find the three-dimensional key points on the initial hand model corresponding to the set position.
  • the execution subject can search for the three-dimensional key points corresponding to the set positions on the initial hand model, so as to realize the matching between the hand image and the initial hand model.
  • Step 304 Adjust the coordinate value of the three-dimensional key point by using the coordinate of the key point to obtain a target hand model corresponding to the hand image.
  • the execution subject can adjust the coordinate values of the three-dimensional key points through the coordinates of the key points, thereby obtaining the target hand model of the hand image. Specifically, the execution subject can adjust the distance between the corresponding three-dimensional key points according to the relative distance between the coordinates of the key points, thereby obtaining the target hand model. In this way, an accurate target hand model can be obtained, the matching degree between the hand image and the target hand model is improved, and the matching accuracy between the hand and the wearing object is improved.
  • Step 305 constructing a virtual wearing object according to the above-mentioned wearing object parameters.
  • step 305 The content of step 305 is the same as that of step 204, and details are not repeated here.
  • Step 306 Match the virtual wearing object with the target hand model to generate a matching image of the virtual wearing object.
  • step 306 The content of step 306 is the same as that of step 205, and details are not repeated here.
  • the virtual wearing object matching method includes the following steps:
  • Step 401 acquiring the image to be processed and the image of the wearing object.
  • step 401 The content of step 401 is the same as that of step 201, and details are not repeated here.
  • Step 402 Identify the above-mentioned image to be processed, and obtain the coordinates of key points of the above-mentioned hand image.
  • step 402 The content of step 402 is the same as that of step 202, and details are not repeated here.
  • Step 403 Adjust the parameters of the initial hand model based on the coordinates of the above-mentioned key points to obtain a target hand model corresponding to the above-mentioned hand image.
  • step 403 The content of step 403 is the same as that of step 203, and details are not repeated here.
  • Step 404 constructing a virtual wearing object according to the above-mentioned wearing object parameters.
  • step 404 The content of step 404 is the same as that of step 204, and details are not repeated here.
  • Step 405 Determine the wearing position corresponding to the virtual wearing item on the target hand space, and set the virtual wearing item corresponding to the wearing position.
  • the wearing object corresponds to the designated wearing position of the hand.
  • rings are worn on the fingers; bracelets are worn on the wrist.
  • the execution subject may determine the wearing position of the target hand space corresponding to the above-mentioned virtual wearing item, and then the executing subject may set the above-mentioned virtual wearing item corresponding to the above-mentioned wearing position in various ways. For example, the execution subject can move the virtual wearing object to the wearing position in three-dimensional space.
  • Step 406 in response to the difference between the outer diameter perimeter of the wearing position and the inner diameter perimeter being less than a set threshold, mark the virtual wearing object as matching the target hand model.
  • the parameters of the wearing object may include the inner diameter perimeter
  • the execution body may measure the outer diameter perimeter of the wearing position and the above-mentioned inner diameter perimeter in the three-dimensional space, and calculate the difference between the two. When the difference is less than the set threshold, it means that the two values are close enough. At this time, the executive body may mark that the virtual wearing object is matched with the target hand model.
  • the above-mentioned matching the above-mentioned virtual wearing object with the above-mentioned target hand model may include: in response to the above-mentioned difference between the outer diameter circumference of the wearing position and the above-mentioned inner diameter circumference greater than Equal to the set threshold, adjust the above wearing parameters.
  • the user wants to get a wearable piece that matches his own hand.
  • the difference between the outer diameter and circumference of the wearing position and the above-mentioned inner diameter and circumference is greater than or equal to the set threshold, it means that the actual size of the wearing object does not match the size of the user's hand wearing position, which may cause the wearer to be too loose or too tight.
  • the execution body can adjust the parameters of the wearing object, so that the difference between the outer diameter perimeter of the wearing position and the above-mentioned inner diameter perimeter is smaller than the set threshold. In this way, targeted adjustment to the user is realized, and the matching effectiveness of the wearing object and the hand is improved.
  • Step 407 Adjust the target hand model after setting the virtual wearing object to the target viewing angle.
  • the hand may be photographed according to a certain spatial angle (ie, a viewing angle).
  • a certain spatial angle ie, a viewing angle
  • the execution subject can adjust the target hand model after setting the virtual wearing object to the target viewing angle.
  • the target angle of view may be the same as the angle of view of the hand image in the image to be processed.
  • the above-mentioned adjusting the above-mentioned target hand model after the above-mentioned virtual wearing object is set to the target angle of view may include the following steps:
  • the first step is to select a set number of keypoint coordinates from the above-mentioned images to be processed as target keypoints.
  • the user usually does not acquire the image to be processed according to the precise three-dimensional space angle (ie, the viewing angle). Therefore, the executive body usually cannot obtain the precise three-dimensional space angle.
  • the execution subject may select a set number of key point coordinates from the above-mentioned to-be-processed images as target key points. Then the target key points can be used to represent the perspective of the hand image in the image to be processed.
  • the second step is to measure the distance information of the above-mentioned target key points on the above-mentioned to-be-processed image.
  • the executive body can measure the distance information of the target key points on the image to be processed.
  • the distance information can be used to determine the structural features (or gesture features) of the hand in space. That is, the distance information has a corresponding relationship with the three-dimensional space angle of the hand. It should be noted that the distance information at this time is the distance between the target key points on the image to be processed, which represents the plane distance between the target key points in the view of the hand image in the image to be processed.
  • the third step is to adjust the target hand model to the target viewing angle based on the distance information.
  • the execution subject can adjust the target hand model based on the distance information, so that the distance between the positions of the target hand model corresponding to the target key points on the plane (not on the three-dimensional space) is the same as the above distance information. In this way, the target hand model can be adjusted to the target viewing angle.
  • Step 408 Generate an initial two-dimensional image corresponding to the target viewing angle.
  • the execution subject can convert the image under the current three-dimensional space target viewing angle into an initial two-dimensional image. Then the hand image in the initial two-dimensional image is approximately the same as the hand image in the to-be-processed image.
  • Step 409 adding the above-mentioned wearing object image in the above-mentioned initial two-dimensional image to the above-mentioned to-be-processed image to obtain a matching image.
  • the display effect of the target hand model is not as real as the display effect of the hand image in the image to be processed.
  • the execution subject may add the above-mentioned wearing object image in the above-mentioned initial two-dimensional image to the above-mentioned to-be-processed image to obtain a matching image. That is, the matching image is a combination of the wearing object image and the image to be processed. In this way, an effect diagram of the wearing object actually being worn on the hand is obtained, which improves the effectiveness of the user selecting the wearing object through the network.
  • the present disclosure provides some embodiments of a virtual wearing object matching device, these device embodiments correspond to those method embodiments shown in FIG. 2 , the The device can be specifically applied to various electronic devices.
  • the virtual wearing object matching device 500 of some embodiments includes: an image acquiring unit 501 , a key point coordinate acquiring unit 502 , a target hand model acquiring unit 503 , a virtual wearing object constructing unit 504 and a matching image generating unit 505 .
  • the image acquisition unit 501 is configured to acquire an image to be processed and an image of the wearing object, the image to be processed includes a hand image, and the wearing object includes parameters of the wearing object;
  • the key point coordinate acquisition unit 502 is configured to obtain the image to be processed.
  • the image is identified, and the key point coordinates of the above-mentioned hand image are obtained;
  • the target hand model obtaining unit 503 is configured to adjust the parameters of the initial hand model based on the above-mentioned key point coordinates to obtain the target hand model corresponding to the above-mentioned hand image;
  • the virtual wearing object construction unit 504 is configured to construct a virtual wearing object according to the above-mentioned wearing object parameters;
  • the matching image generating unit 505 is configured to match the above-mentioned virtual wearing object with the above-mentioned target hand model, and generate the matching of the above-mentioned virtual wearing object image.
  • the above-mentioned image to be processed includes size information, and the above-mentioned size information is used to indicate the size of the actual hand corresponding to the above-mentioned hand image, and,
  • the above-mentioned key point coordinate acquisition unit 502 may include: a designated marker point setting subunit (not shown in the figure) and a first key point coordinate determination subunit (not shown in the figure).
  • the specified marker point setting subunit is configured to set at least one specified marker point corresponding to the set position on the hand image; the first key point coordinate determination subunit is configured to be based on the size information and the at least one specified marker.
  • a designated marker determines the keypoint coordinates of the above hand image.
  • the above-mentioned image to be processed includes point cloud data
  • the above-mentioned point cloud data is used to indicate the three-dimensional coordinates of the hand in space corresponding to the above-mentioned hand image
  • the above-mentioned key point coordinates are obtained.
  • the unit 502 may include: a second key point coordinate determination subunit (not shown in the figure), configured to search for the target point cloud data corresponding to the above-mentioned set position in the above-mentioned point cloud data, and use the coordinates of the above-mentioned target point cloud data. Labeled as keypoint coordinates.
  • the above-mentioned target hand model obtaining unit 503 may include: a three-dimensional key point searching subunit (not shown in the figure) and a target hand model obtaining subunit (not shown in the figure) .
  • the three-dimensional key point search subunit is configured to search for the three-dimensional key point corresponding to the above-mentioned set position on the initial hand model
  • the target hand model acquisition sub-unit is configured to search for the three-dimensional key point by the coordinate of the key point. Adjust the coordinate value of , and obtain the target hand model corresponding to the above hand image.
  • the above-mentioned wearing object parameter includes an inner diameter and circumference
  • the above-mentioned matching image generating unit 505 may include: a position setting subunit (not shown in the figure) and a matching mark subunit (in the figure) not shown).
  • the position setting subunit is configured to determine the wearing position on the target hand space corresponding to the virtual wearing object, and set the virtual wearing object corresponding to the wearing position
  • the matching marking subunit is configured to respond to the wearing position of the wearing position.
  • the difference between the circumference of the outer diameter and the circumference of the inner diameter is smaller than a set threshold value, and is configured to mark that the virtual wearing article is matched with the target hand model.
  • the matching image generation unit 505 may include: a parameter adjustment subunit (not shown in the figure), in response to the difference between the outer diameter circumference of the wearing position and the inner diameter circumference
  • the setting threshold is greater than or equal to, and is configured to adjust the above-mentioned wearing item parameter.
  • the matching image generation unit 505 may include: a target viewing angle adjustment subunit (not shown in the figure), an initial two-dimensional graphics generation subunit (not shown in the figure), and a matching image Generate subunits (not shown in the figure).
  • the target viewing angle adjustment subunit is configured to adjust the target hand model after setting the virtual wearing object to a target viewing angle, and the target viewing angle is the same as the viewing angle of the hand image in the image to be processed;
  • the initial two-dimensional graphic generation The subunit is configured to generate an initial two-dimensional image corresponding to the target viewing angle;
  • the matching image generation subunit is configured to add the wearing object image in the initial two-dimensional image to the to-be-processed image to obtain a matching image.
  • the above-mentioned target viewing angle adjustment subunit may include: a target key point selection module (not shown in the figure), a distance information measurement module (not shown in the figure), and a target viewing angle adjustment module (not shown in the figure).
  • the target key point selection module is configured to select a set number of key point coordinates from the above-mentioned images to be processed as target key points
  • the distance information measurement module is configured to measure the above-mentioned target key points on the above-mentioned image to be processed. distance information
  • a target angle of view adjustment module configured to adjust the above-mentioned target hand model to the target angle of view based on the above-mentioned distance information.
  • the units recorded in the apparatus 500 correspond to the respective steps in the method described with reference to FIG. 2 . Therefore, the operations, features and beneficial effects described above with respect to the method are also applicable to the apparatus 500 and the units included therein, and details are not described herein again.
  • an electronic device 600 may include a processing device (eg, a central processing unit, a graphics processor, etc.) 601 that may be loaded into random access according to a program stored in a read only memory (ROM) 602 or from a storage device 608 Various appropriate actions and processes are executed by the programs in the memory (RAM) 603 . In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored.
  • the processing device 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604.
  • An input/output (I/O) interface 605 is also connected to bus 604 .
  • the following devices can be connected to the I/O interface 605: input devices 606 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; including, for example, a liquid crystal display (LCD), speakers, vibration The output device 607 of the device, etc.; the storage device 608; and the communication device 609.
  • Communication means 609 may allow electronic device 600 to communicate wirelessly or by wire with other devices to exchange data. While FIG. 6 shows electronic device 600 having various means, it should be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in FIG. 6 may represent one device, or may represent multiple devices as required.
  • the processes described above with reference to the flowcharts may be implemented as computer software programs.
  • some embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the method illustrated in the flowchart.
  • the computer program may be downloaded and installed from the network via the communication device 609, or from the storage device 608, or from the ROM 602.
  • the processing device 601 When the computer program is executed by the processing device 601, the above-mentioned functions defined in the methods of some embodiments of the present disclosure are performed.
  • the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the foregoing two.
  • the computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above.
  • a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.
  • a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code therein.
  • Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • a computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device .
  • Program code embodied on a computer readable medium may be transmitted using any suitable medium including, but not limited to, electrical wire, optical fiber cable, RF (radio frequency), etc., or any suitable combination of the foregoing.
  • the client and server can use any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol) to communicate, and can communicate with digital data in any form or medium Communication (eg, a communication network) interconnects.
  • HTTP HyperText Transfer Protocol
  • Examples of communication networks include local area networks (“LAN”), wide area networks (“WAN”), the Internet (eg, the Internet), and peer-to-peer networks (eg, ad hoc peer-to-peer networks), as well as any currently known or future development network of.
  • the above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or may exist alone without being assembled into the electronic device.
  • the above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by the electronic device, the electronic device: acquires an image to be processed and an image of the wearing object, and the above-mentioned image to be processed includes a hand image,
  • the above-mentioned wearing object includes wearing-object parameters; the above-mentioned image to be processed is identified, and the key point coordinates of the above-mentioned hand image are obtained; the parameters of the initial hand model are adjusted based on the above-mentioned key point coordinates, and the target hand model corresponding to the above-mentioned hand image is obtained. ; constructing a virtual wearing object through the wearing object parameters; matching the virtual wearing object with the target hand model to generate a matching image of the virtual wearing object.
  • Computer program code for carrying out operations of some embodiments of the present disclosure may be written in one or more programming languages, including object-oriented programming languages—such as Java, Smalltalk, C++, or a combination thereof, Also included are conventional procedural programming languages - such as the "C" language or similar programming languages.
  • the program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server.
  • the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider to via Internet connection).
  • LAN local area network
  • WAN wide area network
  • each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more logical functions for implementing the specified functions executable instructions.
  • the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations can be implemented in dedicated hardware-based systems that perform the specified functions or operations , or can be implemented in a combination of dedicated hardware and computer instructions.
  • the units described in some embodiments of the present disclosure may be implemented by means of software, and may also be implemented by means of hardware.
  • the described unit can also be set in the processor, for example, it can be described as: a processor includes an image acquisition unit, a key point coordinate acquisition unit, a target hand model acquisition unit, a virtual wearable object construction unit, and a matching image generation unit.
  • a processor includes an image acquisition unit, a key point coordinate acquisition unit, a target hand model acquisition unit, a virtual wearable object construction unit, and a matching image generation unit.
  • the names of these units do not constitute a limitation of the unit itself in some cases, for example, the matching image generating unit may also be described as "a unit for generating matching images".
  • exemplary types of hardware logic components include: Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), Systems on Chips (SOCs), Complex Programmable Logical Devices (CPLDs) and more.
  • FPGAs Field Programmable Gate Arrays
  • ASICs Application Specific Integrated Circuits
  • ASSPs Application Specific Standard Products
  • SOCs Systems on Chips
  • CPLDs Complex Programmable Logical Devices
  • a method for matching a virtual wearing object including: acquiring an image to be processed and an image of a wearing object, the image to be processed includes a hand image, and the wearing object includes wearing object parameters; Recognize the above image to be processed, and obtain the key point coordinates of the above hand image; adjust the parameters of the initial hand model based on the above key point coordinates to obtain a target hand model corresponding to the above hand image; construct a virtual wearer by using the above wearing object parameters object; matching the virtual wearing object with the target hand model to generate a matching image of the virtual wearing object.
  • the image to be processed includes size information, and the size information is used to indicate the actual size of the hand corresponding to the image of the hand, and the image to be processed is identified above, and the obtained image is obtained by recognizing the image to be processed.
  • the key point coordinates of the hand image include: setting at least one designated mark point corresponding to a set position on the hand image; determining the key point coordinates of the hand image based on the size information and the at least one designated mark point.
  • the above-mentioned image to be processed includes point cloud data
  • the above-mentioned point cloud data is used to indicate the three-dimensional coordinates of the hand in space corresponding to the above-mentioned hand image
  • the above-mentioned to-be-processed image Identifying the image to obtain the key point coordinates of the hand image includes: searching for the target point cloud data corresponding to the set position in the point cloud data, and marking the coordinates of the target point cloud data as key point coordinates.
  • adjusting the parameters of the initial hand model based on the coordinates of the key points to obtain the target hand model corresponding to the hand image includes: searching the initial hand model corresponding to the above settings The three-dimensional key point of the position; the coordinate value of the three-dimensional key point is adjusted by the coordinate of the key point, and the target hand model corresponding to the hand image is obtained.
  • the wearing object parameters include an inner diameter and circumference
  • the matching of the virtual wearing object with the target hand model includes: determining that the target hand spatially corresponds to the virtual wearing object the wearing position of the object, and set the virtual wearing object and the wearing position correspondingly; in response to the difference between the outer diameter circumference of the wearing position and the inner diameter circumference being less than the set threshold, mark the virtual wearing object and the target hand. model match.
  • the above-mentioned matching the above-mentioned virtual wearing object with the above-mentioned target hand model includes: in response to the difference between the outer diameter circumference of the wearing position and the above-mentioned inner diameter circumference being greater than or equal to a set threshold value , and adjust the above wearing parameters.
  • generating the matching image of the virtual wearing object includes: adjusting the target hand model after setting the virtual wearing object to a target viewing angle, and the target viewing angle is the same as that in the image to be processed.
  • the angle of view of the hand image is the same; an initial two-dimensional image corresponding to the target angle of view is generated; and the above-mentioned wearing object image in the above-mentioned initial two-dimensional image is added to the above-mentioned image to be processed to obtain a matching image.
  • adjusting the target hand model after setting the virtual wearing object to the target viewing angle includes: selecting a set number of key point coordinates from the to-be-processed image as target key points ; measure the distance information of the above-mentioned target key points on the above-mentioned to-be-processed image; and adjust the above-mentioned target hand model to the target angle of view based on the above-mentioned distance information.
  • a virtual wearing object matching device comprising: an image acquisition unit configured to acquire a to-be-processed image and a wearing object image, the to-be-processed image includes a hand image, the above-mentioned wearing The object includes wearing parameters; the key point coordinate acquisition unit is configured to recognize the above-mentioned image to be processed, and acquire the key point coordinates of the above-mentioned hand image; the target hand model acquisition unit is configured to adjust the initial value based on the above-mentioned key point coordinates.
  • a virtual wearing object construction unit is configured to construct a virtual wearing object by using the above-mentioned wearing object parameters
  • a matching image generation unit is configured to convert the above-mentioned virtual wearing object Matching with the target hand model to generate a matching image of the virtual wearing object.
  • the image to be processed includes size information
  • the size information is used to indicate the size of the actual hand corresponding to the hand image
  • the key point coordinate acquisition unit includes: specifying a marker point
  • the setting subunit is configured to set at least one designated marker point corresponding to the set position on the above-mentioned hand image
  • the first key point coordinate determination subunit is configured to determine the above-mentioned size information and the above-mentioned at least one designated marker point based on the above-mentioned size information Keypoint coordinates of the hand image.
  • the image to be processed includes point cloud data
  • the point cloud data is used to indicate the three-dimensional coordinates of the hand in space corresponding to the hand image
  • the key point coordinates are obtained
  • the unit includes: a second key point coordinate determination subunit, configured to search for target point cloud data corresponding to the set position in the point cloud data, and mark the coordinates of the target point cloud data as key point coordinates.
  • the above-mentioned target hand model acquisition unit includes: a three-dimensional key point search subunit configured to search for three-dimensional key points on the above-mentioned initial hand model corresponding to the above-mentioned set positions;
  • the model obtaining subunit is configured to adjust the coordinate value of the three-dimensional key point according to the coordinate of the key point, so as to obtain the target hand model corresponding to the hand image.
  • the wearing object parameter includes an inner diameter and circumference
  • the matching image generating unit includes: a position setting subunit configured to determine the target hand space corresponding to the virtual wearing object. wearing position, and set the above-mentioned virtual wearing object corresponding to the above-mentioned wearing position; the matching marking subunit is configured to mark the above-mentioned virtual wearing in response to the difference between the outer diameter perimeter of the above-mentioned wearing position and the above-mentioned inner diameter perimeter being less than the set threshold
  • the object matches the target hand model above.
  • the matching image generation unit includes: a parameter adjustment subunit configured to adjust the difference between the outer circumference of the wearing position and the inner circumference of the wearing position is greater than or equal to a set threshold. The above wearing parameters.
  • the matching image generating unit includes: a target viewing angle adjustment subunit configured to adjust the target hand model after setting the virtual wearing object to a target viewing angle, the target viewing angle is the same as the above The angle of view of the hand image in the image to be processed is the same; the initial two-dimensional image generation subunit is configured to generate an initial two-dimensional image corresponding to the above-mentioned target angle of view; the matching image generation subunit is configured to The above-mentioned wearing object image is added to the above-mentioned to-be-processed image to obtain a matching image.
  • the target viewing angle adjustment subunit includes: a target key point selection module configured to select a set number of key point coordinates from the above-mentioned to-be-processed image as target key points; distance information measurement The module is configured to measure the distance information of the target key points on the image to be processed; the target viewing angle adjustment module is configured to adjust the target hand model to the target viewing angle based on the distance information.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Human Computer Interaction (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Processing Or Creating Images (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

本公开的实施例公开了虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质。该方法的一具体实施方式包括：获取待处理图像和佩戴物图像，上述待处理图像包括手部图像，上述佩戴物图像包括佩戴物参数；对上述待处理图像进行识别，获取上述手部图像的关键点坐标；基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型；通过上述佩戴物参数构建虚拟佩戴物；将上述虚拟佩戴物与上述目标手部模型进行匹配，生成上述虚拟佩戴物的匹配图像。该实施方式实现了手部与佩戴物的虚拟匹配的准确性和有效性，提高了用户通过网络选择物品的有效性。

Description

虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质
相关申请的交叉引用
本申请基于申请号为202011095497.1、申请日为2020年10月14日，名称为“虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质”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
本公开的实施例涉及计算机技术领域，具体涉及虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质。
背景技术
随着信息技术的发展，越来越多的物品通过网络进行展示。用户可以通过网络浏览物品，并选择需要的物品。
对于某些普适性的物品(例如可以是手机、电脑等)，不论男女老幼，用户都可以根据需要直接选择。而对于某些对使用有个性化要求的物品(例如可以是戒指、手镯、手表等佩戴物)等，用户需要通过实际佩戴才能确定是否合适。当前网络无法实现用户对此类物品的佩戴需求，无法为用户提供个性化服务。
发明内容
本公开的内容部分用于以简要的形式介绍构思，这些构思将在后面的具体实施方式部分被详细描述。本公开的内容部分并不旨在标识要求保护的技术方案的关键特征或必要特征，也不旨在用于限制所要求的保护的技术方案的范围。
本公开的一些实施例提出了虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质，来解决以上背景技术部分提到的技术问题。
第一方面，本公开的一些实施例提供了一种虚拟佩戴物匹配方法，该 方法包括：获取待处理图像和佩戴物图像，上述待处理图像包括手部图像，上述佩戴物图像包括佩戴物参数；对上述待处理图像进行识别，获取上述手部图像的关键点坐标；基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型；通过上述佩戴物参数构建虚拟佩戴物；将上述虚拟佩戴物与上述目标手部模型进行匹配，生成上述虚拟佩戴物的匹配图像。
第二方面，本公开的一些实施例提供了一种虚拟佩戴物匹配装置，该装置包括：图像获取单元，被配置成获取待处理图像和佩戴物图像，上述待处理图像包括手部图像，上述佩戴物图像包括佩戴物参数；关键点坐标获取单元，被配置成对上述待处理图像进行识别，获取上述手部图像的关键点坐标；目标手部模型获取单元，被配置成基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型；虚拟佩戴物构建单元，被配置成通过上述佩戴物参数构建虚拟佩戴物；匹配图像生成单元，被配置成将上述虚拟佩戴物与上述目标手部模型进行匹配，生成上述虚拟佩戴物的匹配图像。
第三方面，本公开的一些实施例提供了一种电子设备，包括：一个或多个处理器；存储器，其上存储有一个或多个程序，当上述一个或多个程序被上述一个或多个处理器执行时，使得上述一个或多个处理器执行上述第一方面的虚拟佩戴物匹配方法。
第四方面，本公开的一些实施例提供了一种计算机可读介质，其上存储有计算机程序，其特征在于，该程序被处理器执行时实现上述第一方面的虚拟佩戴物匹配方法。
本公开的上述各个实施例中的一个实施例具有如下有益效果：首先获取待处理图像和佩戴物图像，并确定待处理图像中手部图像的关键点坐标，确定了手部的参数；然后通过关键点坐标调整初始手部模型的参数，得到目标手部模型，实现了手部的虚拟化；之后再通过佩戴物参数构建虚拟佩戴物，实现了佩戴物虚拟化；最后对虚拟佩戴物和目标手部模型进行匹配，生成匹配图像。实现了手部与佩戴物的虚拟匹配的准确性和有效性，提高了用户通过网络选择物品的有效性。
附图说明
结合附图并参考以下具体实施方式，本公开各实施例的上述和其他特征、优点及方面将变得更加明显。贯穿附图中，相同或相似的附图标记表示相同或相似的元素。应当理解附图是示意性的，元件和元素不一定按照比例绘制。
图1是本公开的一些实施例的虚拟佩戴物匹配方法的应用场景的示意图；
图2是根据本公开的虚拟佩戴物匹配方法的一些实施例的流程图；
图3是根据本公开的虚拟佩戴物匹配方法的另一些实施例的流程图；
图4是根据本公开的虚拟佩戴物匹配方法的又一些实施例的流程图；
图5是根据本公开的虚拟佩戴物匹配装置的一些实施例的结构示意图；
图6是适于用来实现本公开的一些实施例的电子设备的结构示意图。
具体实施方式
下面将参照附图更详细地描述本公开的实施例。虽然附图中显示了本公开的某些实施例，然而应当理解的是，本公开可以通过各种形式来实现，而且不应该被解释为限于这里阐述的实施例。相反，提供这些实施例是为了更加透彻和完整地理解本公开。应当理解的是，本公开的附图及实施例仅用于示例性作用，并非用于限制本公开的保护范围。
另外还需要说明的是，为了便于描述，附图中仅示出了与有关发明相关的部分。在不冲突的情况下，本公开中的实施例及实施例中的特征可以相互组合。
需要注意，本公开中提及的“第一”、“第二”等概念仅用于对不同的装置、模块或单元进行区分，并非用于限定这些装置、模块或单元所执行的功能的顺序或者相互依存关系。
需要注意，本公开中提及的“一个”、“多个”的修饰是示意性而非限制性的，本领域技术人员应当理解，除非在上下文另有明确指出，否则应该理解为“一个或多个”。
本公开实施方式中的多个装置之间所交互的消息或者信息的名称仅用于说明性的目的，而并不是用于对这些消息或信息的范围进行限制。
下面将参考附图并结合实施例来详细说明本公开。
图1是根据本公开一些实施例的虚拟佩戴物匹配方法的一个应用场景的示意图。
如图1所示，电子设备101(例如可以是佩戴物匹配服务器)获取到包含手部图像的待处理图像和戒指图像(即佩戴物图像)后，首先对该待处理图像进行识别，以获取待处理图像中手部图像的关键点坐标。其中，关键点可以是手部的多个指定特征点(例如可以是手指关节处)。关键点坐标可以是以待处理图像中设定点为原点(例如可以是待处理图像的坐下角)构建坐标平面后，各个关键点在坐标平面上的坐标。需要说明的是，本公开的关键点坐标可以用于表征对应上述手部图像的手部的设定位置之间的相对位置关系。例如，某一手指上有3个关键点，则对应这3个关键点的关键点坐标之间的相对距离(即相对位置关系)就可以表征该手指的长度等信息。然后，电子设备101可以根据获取的关键点坐标调整预先构建的初始手部模型的参数，使得修改参数后的目标手部模型的各个参数与手部图像对应的实际手部的参数相同。之后，电子设备101还可以通过戒指参数构建戒指模型，并对戒指模型进行渲染，得到虚拟戒指。最后，电子设备101可以将虚拟戒指与目标手部模型进行匹配。当匹配成功后，电子设备101可以生成匹配图像。如此，实现了戒指与手部的虚拟匹配，提高了用户通过网络选择物品的有效性。
应该理解，图1中的电子设备数目仅仅是示意性的。根据实现需要，可以具有任意数目的终端设备。
继续参考图2，示出了根据本公开的虚拟佩戴物匹配方法的一些实施例的流程200。该虚拟佩戴物匹配方法，包括以下步骤：
步骤201，获取待处理图像和佩戴物图像。
在一些实施例中，虚拟佩戴物匹配方法的执行主体(例如图1所示的服务器101)可以通过有线连接方式或者无线连接方式获取待处理图像和佩戴物图像。其中，佩戴物图像可以是用户从网络上获取的图像，通常为用户期望佩戴的物体图像。待处理图像可以为用户通过各种图像采集装置(例如可以是智能手机)获取的图像。需要指出的是，上述无线连接方式可以包括但不限于3G/4G连接、WiFi连接、蓝牙连接、WiMAX连接、Zigbee 连接、UWB(ultra wideband)连接、以及其他现在已知或将来开发的无线连接方式。
当用户在网络上浏览物品时，想要知道选择的物品是否适合时，可以同时将包含手部图像的待处理图像和佩戴物图像发送给执行主体。为了确定佩戴物是否适合，佩戴物图像可以包括佩戴物参数。佩戴物参数可以是用户从网络上对佩戴物的描述中获取的信息。
步骤202，对上述待处理图像进行识别，获取上述手部图像的关键点坐标。
执行主体可以从待处理图像中识别出手部图像。然后，执行主体可以以待处理图像上的某一个点为原点构建平面坐标系，然后执行主体可以查找手部图像上的关键点，并确定关键点在平面坐标系中的坐标。
步骤203，基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型。
执行主体可以通过关键点坐标确定手部图像中手部各个关键点之间的距离关系。然后，执行主体可以通过关键点坐标来调整初始手部模型的参数，使得初始手部模型尽量与手部图像对应的实际手部尺寸相同，有利于提高佩戴物与手部匹配的准确性。
步骤204，通过上述佩戴物参数构建虚拟佩戴物。
为了实现佩戴物与手部的匹配，执行主体还可以通过佩戴物参数构建佩戴物模型。然后，执行主体通过渲染等方式得到虚拟佩戴物。
步骤205，将上述虚拟佩戴物与上述目标手部模型进行匹配，生成上述虚拟佩戴物的匹配图像。
得到虚拟佩戴物和目标手部模型后，完成了将手部图像和佩戴物的虚拟化处理。虚拟佩戴物和目标手部模型在尺寸和形态等方面分别与佩戴物和手部相同。执行主体可以将虚拟佩戴物与上述目标手部模型进行匹配。当虚拟佩戴物与目标手部模型匹配成功时，执行主体可以生成上述虚拟佩戴物的匹配图像。如此，实现了佩戴物与手部的虚拟匹配，提高了用户通过网络选择物品的有效性。
本公开的一些实施例公开的虚拟佩戴物匹配方法，首先获取待处理图像和佩戴物图像，并确定待处理图像中手部图像的关键点坐标，确定了手部 的参数；然后通过关键点坐标调整初始手部模型的参数，得到目标手部模型，实现了手部的虚拟化；之后再通过佩戴物参数构建虚拟佩戴物，实现了佩戴物虚拟化；最后对虚拟佩戴物和目标手部模型进行匹配，生成匹配图像。实现了手部与佩戴物的虚拟匹配的准确性和有效性，提高了用户通过网络选择物品的有效性。
继续参考图3，示出了根据本公开的虚拟佩戴物匹配方法的一些实施例的流程300。该虚拟佩戴物匹配方法，包括以下步骤：
步骤301，获取待处理图像和佩戴物图像。
步骤301的内容与步骤201的内容相同，此处不再一一赘述。
步骤302，对上述待处理图像进行识别，获取上述手部图像的关键点坐标。
步骤302的内容与步骤202的内容相同，此处不再一一赘述。
在一些实施例的一些可选的实现方式中，上述对上述待处理图像进行识别，获取上述手部图像的关键点坐标，可以包括以下步骤：
第一步，在上述手部图像上设置对应上述设定位置的至少一个指定标记点。
执行主体可以根据初始手部模型的需要，在手部图像上设置至少一个指定标记点。其中设定位置可以为手部各个关节点。关节点可以表征手部的各个尺寸，有利于提高目标手部模型的精度。
第二步，基于尺寸信息和上述至少一个指定标记点确定上述手部图像的关键点坐标。
待处理图像还可以包括尺寸信息。其中，上述尺寸信息可以用于指示上述手部图像对应的实际手部的尺寸。即，尺寸信息可以是用户通过测量手部的实际尺寸得到的信息。例如，用户想购买戒指，在向执行主体发送待处理图像时，可以一并发送用户手部的尺寸信息。尺寸信息可以是：手部的长度、各个手指的长度、手指的宽度、手掌的厚度等。通常，尺寸信息较为准确，指定标记点标记时的误差相对较大。执行主体可以通过尺寸信息和指定标记点结合起来确定关键点坐标。如此，提高了获取关键点坐标的准确性。
在一些实施例的一些可选的实现方式中，上述对上述待处理图像进行 识别，获取上述手部图像的关键点坐标，可以包括：查找上述点云数据中对应上述设定位置的目标点云数据，并将上述目标点云数据的坐标标记为关键点坐标。
当用户获取的待处理图像为通过点云数据采集设备获取的三维图像时，待处理图像可以包括点云数据。其中，上述点云数据用于指示上述手部图像对应的手部在空间中的三维坐标。此时，执行主体可以直接查询对应设定位置的目标点云数据，并将目标点云数据的坐标标记为关键点坐标。此时的关键点坐标为三维坐标，可以更为方便地调整初始手部模型的参数。
步骤303，查找上述初始手部模型上对应上述设定位置的三维关键点。
执行主体可以查找初始手部模型上对应设定位置的三维关键点，以实现手部图像与初始手部模型的匹配。
步骤304，通过上述关键点坐标对上述三维关键点的坐标值进行调整，得到对应上述手部图像的目标手部模型。
得到关键点坐标后，执行主体可以通过关键点坐标对三维关键点的坐标值进行调整，进而得到手部图像的目标手部模型。具体的，执行主体可以根据关键点坐标之间的相对距离来调整对应的三维关键点之间的距离，进而得到目标手部模型。如此，可以得到精确的目标手部模型，提高了手部图像与目标手部模型的匹配度，有利于提高手部与佩戴物的匹配精度。
步骤305，通过上述佩戴物参数构建虚拟佩戴物。
步骤305的内容与步骤204的内容相同，此处不再一一赘述。
步骤306，将上述虚拟佩戴物与上述目标手部模型进行匹配，生成上述虚拟佩戴物的匹配图像。
步骤306的内容与步骤205的内容相同，此处不再一一赘述。
继续参考图4，示出了根据本公开的虚拟佩戴物匹配方法的一些实施例的流程400。该虚拟佩戴物匹配方法，包括以下步骤：
步骤401，获取待处理图像和佩戴物图像。
步骤401的内容与步骤201的内容相同，此处不再一一赘述。
步骤402，对上述待处理图像进行识别，获取上述手部图像的关键点坐标。
步骤402的内容与步骤202的内容相同，此处不再一一赘述。
步骤403，基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型。
步骤403的内容与步骤203的内容相同，此处不再一一赘述。
步骤404，通过上述佩戴物参数构建虚拟佩戴物。
步骤404的内容与步骤204的内容相同，此处不再一一赘述。
步骤405，确定上述目标手部空间上对应上述虚拟佩戴物的佩戴位置，并将上述虚拟佩戴物与上述佩戴位置对应设置。
通常，佩戴物与手部的指定佩戴位置对应。例如，戒指佩戴在手指上；手镯佩戴在手腕处。执行主体可以确定目标手部空间上对应上述虚拟佩戴物的佩戴位置，然后执行主体可以通过多种方式将上述虚拟佩戴物与上述佩戴位置对应设置。例如，执行主体可以在三维空间里移动虚拟佩戴物至佩戴位置。
步骤406，响应于上述佩戴位置的外径周长与上述内径周长的差值小于设置阈值，标记上述虚拟佩戴物与上述目标手部模型匹配。
实际中，佩戴物与手部是否匹配，可以通过佩戴物的内径与佩戴位置的外径来衡量。具体的，佩戴物参数可以包括内径周长，执行主体可以在三维空间里测量佩戴位置的外径周长和上述内径周长，并计算两者的差值。当差值小于设定阈值时，说明两个的数值足够接近。此时，执行主体可以标记上述虚拟佩戴物与上述目标手部模型匹配。
在一些实施例的一些可选的实现方式中，上述将上述虚拟佩戴物与上述目标手部模型进行匹配，可以包括：响应于上述佩戴位置的外径周长与上述内径周长的差值大于等于设置阈值，调整上述佩戴物参数。
通常，用户想得到和自身手部相匹配的佩戴物。当佩戴位置的外径周长与上述内径周长的差值大于等于设置阈值时，说明佩戴物的实际尺寸与用户的手部的佩戴位置的尺寸不匹配，可能出现佩戴过松或过紧的情况。为了适应用户的实际需求，执行主体可以调整佩戴物参数，使得佩戴位置的外径周长与上述内径周长的差值小于设置阈值。如此，实现了对用户的针对性调整，提高了佩戴物与手部的匹配有效性。
步骤407，将设置上述虚拟佩戴物后的上述目标手部模型调整至目标 视角。
实际中，由于用户自己的喜好等原因，用户在获取待处理图像时，手部可以是按照一定的空间角度(即视角)拍摄的。为了使得用户得到虚拟环境下佩戴物与手部的效果图，执行主体可以将设置虚拟佩戴物后的目标手部模型调整至目标视角。其中，上述目标视角可以是与上述待处理图像中手部图像的视角相同。
在一些实施例的一些可选的实现方式中，上述将设置上述虚拟佩戴物后的上述目标手部模型调整至目标视角，可以包括以下步骤：
第一步，从上述待处理图像中选择设定数量的关键点坐标作为目标关键点。
实际中，用户通常不是按照精确的三维空间角度(即视角)来获取待处理图像的。因此，执行主体通常无法获取精确的三维空间角度。此时，执行主体可以从上述待处理图像中选择设定数量的关键点坐标作为目标关键点。则目标关键点可以用来表征待处理图像中手部图像的视角。
第二步，测量上述目标关键点在上述待处理图像上的距离信息。
确定了目标关键点后，执行主体可以测量目标关键点在待处理图像上的距离信息。距离信息可以用于确定手部在空间的结构特征(或姿态特征)。即，距离信息与手部的三维空间角度具有对应关系。需要说明的是，此时的距离信息为待处理图像上的目标关键点之间的距离，表征的是待处理图像中手部图像的视角下目标关键点之间的平面距离。
第三步，基于上述距离信息将上述目标手部模型调整至目标视角。
执行主体可以基于距离信息对目标手部模型进行调整，使得目标手部模型上对应目标关键点的位置之间，在平面上(不是三维空间上)的距离与上述的距离信息对应相同。如此，可以将目标手部模型调整为目标视角。
步骤408，生成对应上述目标视角的初始二维图像。
确定了目标视角后，执行主体可以将当前的三维空间目标视角下的图像转换为初始二维图像。则初始二维图像中的手部图像与待处理图像中手部图像近似相同。
步骤409，将上述初始二维图像中的上述佩戴物图像添加至上述待处理图像，得到匹配图像。
实际中，目标手部模型的显示效果没有待处理图像中的手部图像的显示效果真实。为此，执行主体可以将上述初始二维图像中的上述佩戴物图像添加至上述待处理图像，得到匹配图像。即，匹配图像为佩戴物图像和待处理图像的结合。如此，得到了佩戴物实际佩戴在手部的效果图，提高了用户通过网络选择佩戴物的有效性。
进一步参考图5，作为对上述各图所示方法的实现，本公开提供了一种虚拟佩戴物匹配装置的一些实施例，这些装置实施例与图2所示的那些方法实施例相对应，该装置具体可以应用于各种电子设备中。
如图5所示，一些实施例的虚拟佩戴物匹配装置500包括：图像获取单元501、关键点坐标获取单元502、目标手部模型获取单元503、虚拟佩戴物构建单元504和匹配图像生成单元505。其中，图像获取单元501，被配置成获取待处理图像和佩戴物图像，上述待处理图像包括手部图像，上述佩戴物包括佩戴物参数；关键点坐标获取单元502，被配置成对上述待处理图像进行识别，获取上述手部图像的关键点坐标；目标手部模型获取单元503，被配置成基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型；虚拟佩戴物构建单元504，被配置成通过上述佩戴物参数构建虚拟佩戴物；匹配图像生成单元505，被配置成将上述虚拟佩戴物与上述目标手部模型进行匹配，生成上述虚拟佩戴物的匹配图像。
在一些实施例的可选实现方式中，上述待处理图像包括尺寸信息，上述尺寸信息用于指示上述手部图像对应的实际手部的尺寸，以及，
上述关键点坐标获取单元502可以包括：指定标记点设置子单元(图中未示出)和第一关键点坐标确定子单元(图中未示出)。其中，指定标记点设置子单元，被配置成在上述手部图像上设置对应上述设定位置的至少一个指定标记点；第一关键点坐标确定子单元，被配置成基于上述尺寸信息和上述至少一个指定标记点确定上述手部图像的关键点坐标。
在一些实施例的可选实现方式中，上述待处理图像包括点云数据，上述点云数据用于指示上述手部图像对应的手部的在空间中的三维坐标，以及，上述关键点坐标获取单元502可以包括：第二关键点坐标确定子单元(图中未示出)，被配置成查找上述点云数据中对应上述设定位置的目标 点云数据，并将上述目标点云数据的坐标标记为关键点坐标。
在一些实施例的可选实现方式中，上述目标手部模型获取单元503可以包括：三维关键点查找子单元(图中未示出)和目标手部模型获取子单元(图中未示出)。其中，三维关键点查找子单元，被配置成查找上述初始手部模型上对应上述设定位置的三维关键点；目标手部模型获取子单元，被配置成通过上述关键点坐标对上述三维关键点的坐标值进行调整，得到对应上述手部图像的目标手部模型。
在一些实施例的可选实现方式中，上述佩戴物参数包括内径周长，以及，上述匹配图像生成单元505可以包括：位置设置子单元(图中未示出)和匹配标记子单元(图中未示出)。其中，位置设置子单元，被配置成确定上述目标手部空间上对应上述虚拟佩戴物的佩戴位置，并将上述虚拟佩戴物与上述佩戴位置对应设置；匹配标记子单元，响应于上述佩戴位置的外径周长与上述内径周长的差值小于设置阈值，被配置成标记上述虚拟佩戴物与上述目标手部模型匹配。
在一些实施例的可选实现方式中，上述匹配图像生成单元505可以包括：参数调整子单元(图中未示出)，响应于上述佩戴位置的外径周长与上述内径周长的差值大于等于设置阈值，被配置成调整上述佩戴物参数。
在一些实施例的可选实现方式中，上述匹配图像生成单元505可以包括：目标视角调整子单元(图中未示出)、初始二维图形生成子单元(图中未示出)和匹配图像生成子单元(图中未示出)。其中，目标视角调整子单元，被配置成将设置上述虚拟佩戴物后的上述目标手部模型调整至目标视角，上述目标视角与上述待处理图像中手部图像的视角相同；初始二维图形生成子单元，被配置成生成对应上述目标视角的初始二维图像；匹配图像生成子单元，被配置成将上述初始二维图像中的上述佩戴物图像添加至上述待处理图像，得到匹配图像。
在一些实施例的可选实现方式中，上述目标视角调整子单元可以包括：目标关键点选择模块(图中未示出)、距离信息测量模块(图中未示出)和目标视角调整模块(图中未示出)。其中，目标关键点选择模块，被配置成从上述待处理图像中选择设定数量的关键点坐标作为目标关键点；距离信息测量模块，被配置成测量上述目标关键点在上述待处理图像 上的距离信息；目标视角调整模块，被配置成基于上述距离信息将上述目标手部模型调整至目标视角。
可以理解的是，该装置500中记载的诸单元与参考图2描述的方法中的各个步骤相对应。由此，上文针对方法描述的操作、特征以及产生的有益效果同样适用于装置500及其中包含的单元，在此不再赘述。
如图6所示，电子设备600可以包括处理装置(例如中央处理器、图形处理器等)601，其可以根据存储在只读存储器(ROM)602中的程序或者从存储装置608加载到随机访问存储器(RAM)603中的程序而执行各种适当的动作和处理。在RAM 603中，还存储有电子设备600操作所需的各种程序和数据。处理装置601、ROM 602以及RAM 603通过总线604彼此相连。输入/输出(I/O)接口605也连接至总线604。
通常，以下装置可以连接至I/O接口605：包括例如触摸屏、触摸板、键盘、鼠标、摄像头、麦克风、加速度计、陀螺仪等的输入装置606；包括例如液晶显示器(LCD)、扬声器、振动器等的输出装置607；存储装置608；以及通信装置609。通信装置609可以允许电子设备600与其他设备进行无线或有线通信以交换数据。虽然图6示出了具有各种装置的电子设备600，但是应理解的是，并不要求实施或具备所有示出的装置。可以替代地实施或具备更多或更少的装置。图6中示出的每个方框可以代表一个装置，也可以根据需要代表多个装置。
特别地，根据本公开的一些实施例，上文参考流程图描述的过程可以被实现为计算机软件程序。例如，本公开的一些实施例包括一种计算机程序产品，其包括承载在计算机可读介质上的计算机程序，该计算机程序包含用于执行流程图所示的方法的程序代码。在这样的一些实施例中，该计算机程序可以通过通信装置609从网络上被下载和安装，或者从存储装置608被安装，或者从ROM 602被安装。在该计算机程序被处理装置601执行时，执行本公开的一些实施例的方法中限定的上述功能。
需要说明的是，本公开的一些实施例上述的计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质或者是上述两者的任意组合。计算机可读存储介质例如可以是——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。计算机可读 存储介质的更具体的例子可以包括但不限于：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本公开的一些实施例中，计算机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。而在本公开的一些实施例中，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，其中承载了计算机可读的程序代码。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。计算机可读信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读信号介质可以发送、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。计算机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：电线、光缆、RF(射频)等等，或者上述的任意合适的组合。
在一些实施方式中，客户端、服务器可以利用诸如HTTP(HyperText Transfer Protocol，超文本传输协议)之类的任何当前已知或未来研发的网络协议进行通信，并且可以与任意形式或介质的数字数据通信(例如，通信网络)互连。通信网络的示例包括局域网(“LAN”)，广域网(“WAN”)，网际网(例如，互联网)以及端对端网络(例如，ad hoc端对端网络)，以及任何当前已知或未来研发的网络。
上述计算机可读介质可以是上述电子设备中所包含的；也可以是单独存在，而未装配入该电子设备中。上述计算机可读介质承载有一个或者多个程序，当上述一个或者多个程序被该电子设备执行时，使得该电子设备：获取待处理图像和佩戴物图像，上述待处理图像包括手部图像，上述佩戴物包括佩戴物参数；对上述待处理图像进行识别，获取上述手部图像的关键点坐标；基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型；通过上述佩戴物参数构建虚拟佩戴物；将上述虚拟佩戴物与上述目标手部模型进行匹配，生成上述虚拟佩戴物的匹配图像。
可以以一种或多种程序设计语言或其组合来编写用于执行本公开的一些实施例的操作的计算机程序代码，上述程序设计语言包括面向对象的程序设计语言—诸如Java、Smalltalk、C++，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机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络——包括局域网(LAN)或广域网(WAN)——连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。
附图中的流程图和框图，图示了按照本公开各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，该模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
描述于本公开的一些实施例中的单元可以通过软件的方式实现，也可以通过硬件的方式来实现。所描述的单元也可以设置在处理器中，例如，可以描述为：一种处理器包括图像获取单元、关键点坐标获取单元、目标手部模型获取单元、虚拟佩戴物构建单元和匹配图像生成单元。其中，这些单元的名称在某种情况下并不构成对该单元本身的限定，例如，匹配图像生成单元还可以被描述为“用于生成匹配图像的单元”。
本文中以上描述的功能可以至少部分地由一个或多个硬件逻辑部件来执行。例如，非限制性地，可以使用的示范类型的硬件逻辑部件包括：现场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准产品(ASSP)、 片上系统(SOC)、复杂可编程逻辑设备(CPLD)等等。
根据本公开的一个或多个实施例，提供了一种虚拟佩戴物匹配方法，包括：获取待处理图像和佩戴物图像，上述待处理图像包括手部图像，上述佩戴物包括佩戴物参数；对上述待处理图像进行识别，获取上述手部图像的关键点坐标；基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型；通过上述佩戴物参数构建虚拟佩戴物；将上述虚拟佩戴物与上述目标手部模型进行匹配，生成上述虚拟佩戴物的匹配图像。
根据本公开的一个或多个实施例，上述待处理图像包括尺寸信息，上述尺寸信息用于指示上述手部图像对应的实际手部的尺寸，以及，上述对上述待处理图像进行识别，获取上述手部图像的关键点坐标，包括：在上述手部图像上设置对应设定位置的至少一个指定标记点；基于上述尺寸信息和上述至少一个指定标记点确定上述手部图像的关键点坐标。
根据本公开的一个或多个实施例，上述待处理图像包括点云数据，上述点云数据用于指示上述手部图像对应的手部的在空间中的三维坐标，以及，上述对上述待处理图像进行识别，获取上述手部图像的关键点坐标，包括：查找上述点云数据中对应上述设定位置的目标点云数据，并将上述目标点云数据的坐标标记为关键点坐标。
根据本公开的一个或多个实施例，上述基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型，包括：查找上述初始手部模型上对应上述设定位置的三维关键点；通过上述关键点坐标对上述三维关键点的坐标值进行调整，得到对应上述手部图像的目标手部模型。
根据本公开的一个或多个实施例，上述佩戴物参数包括内径周长，以及，上述将上述虚拟佩戴物与上述目标手部模型进行匹配，包括：确定上述目标手部空间上对应上述虚拟佩戴物的佩戴位置，并将上述虚拟佩戴物与上述佩戴位置对应设置；响应于上述佩戴位置的外径周长与上述内径周长的差值小于设置阈值，标记上述虚拟佩戴物与上述目标手部模型匹配。
根据本公开的一个或多个实施例，上述将上述虚拟佩戴物与上述目标手部模型进行匹配，包括：响应于上述佩戴位置的外径周长与上述内径周 长的差值大于等于设置阈值，调整上述佩戴物参数。
根据本公开的一个或多个实施例，上述生成上述虚拟佩戴物的匹配图像，包括：将设置上述虚拟佩戴物后的上述目标手部模型调整至目标视角，上述目标视角与上述待处理图像中手部图像的视角相同；生成对应上述目标视角的初始二维图像；将上述初始二维图像中的上述佩戴物图像添加至上述待处理图像，得到匹配图像。
根据本公开的一个或多个实施例，上述将设置上述虚拟佩戴物后的上述目标手部模型调整至目标视角，包括：从上述待处理图像中选择设定数量的关键点坐标作为目标关键点；测量上述目标关键点在上述待处理图像上的距离信息；基于上述距离信息将上述目标手部模型调整至目标视角。
根据本公开的一个或多个实施例，提供了一种虚拟佩戴物匹配装置，包括：图像获取单元，被配置成获取待处理图像和佩戴物图像，上述待处理图像包括手部图像，上述佩戴物包括佩戴物参数；关键点坐标获取单元，被配置成对上述待处理图像进行识别，获取上述手部图像的关键点坐标；目标手部模型获取单元，被配置成基于上述关键点坐标调整初始手部模型的参数，得到对应上述手部图像的目标手部模型；虚拟佩戴物构建单元，被配置成通过上述佩戴物参数构建虚拟佩戴物；匹配图像生成单元，被配置成将上述虚拟佩戴物与上述目标手部模型进行匹配，生成上述虚拟佩戴物的匹配图像。
根据本公开的一个或多个实施例，上述待处理图像包括尺寸信息，上述尺寸信息用于指示上述手部图像对应的实际手部的尺寸，以及，上述关键点坐标获取单元包括：指定标记点设置子单元，被配置成在上述手部图像上设置对应设定位置的至少一个指定标记点；第一关键点坐标确定子单元，被配置成基于上述尺寸信息和上述至少一个指定标记点确定上述手部图像的关键点坐标。
根据本公开的一个或多个实施例，上述待处理图像包括点云数据，上述点云数据用于指示上述手部图像对应的手部的在空间中的三维坐标，以及，上述关键点坐标获取单元包括：第二关键点坐标确定子单元，被配置成查找上述点云数据中对应上述设定位置的目标点云数据，并将上述目标点云数据的坐标标记为关键点坐标。
根据本公开的一个或多个实施例，上述目标手部模型获取单元包括：三维关键点查找子单元，被配置成查找上述初始手部模型上对应上述设定位置的三维关键点；目标手部模型获取子单元，被配置成通过上述关键点坐标对上述三维关键点的坐标值进行调整，得到对应上述手部图像的目标手部模型。
根据本公开的一个或多个实施例，上述佩戴物参数包括内径周长，以及，上述匹配图像生成单元包括：位置设置子单元，被配置成确定上述目标手部空间上对应上述虚拟佩戴物的佩戴位置，并将上述虚拟佩戴物与上述佩戴位置对应设置；匹配标记子单元，响应于上述佩戴位置的外径周长与上述内径周长的差值小于设置阈值，被配置成标记上述虚拟佩戴物与上述目标手部模型匹配。
根据本公开的一个或多个实施例，上述匹配图像生成单元包括：参数调整子单元，响应于上述佩戴位置的外径周长与上述内径周长的差值大于等于设置阈值，被配置成调整上述佩戴物参数。
根据本公开的一个或多个实施例，上述匹配图像生成单元包括：目标视角调整子单元，被配置成将设置上述虚拟佩戴物后的上述目标手部模型调整至目标视角，上述目标视角与上述待处理图像中手部图像的视角相同；初始二维图形生成子单元，被配置成生成对应上述目标视角的初始二维图像；匹配图像生成子单元，被配置成将上述初始二维图像中的上述佩戴物图像添加至上述待处理图像，得到匹配图像。
根据本公开的一个或多个实施例，上述目标视角调整子单元包括：目标关键点选择模块，被配置成从上述待处理图像中选择设定数量的关键点坐标作为目标关键点；距离信息测量模块，被配置成测量上述目标关键点在上述待处理图像上的距离信息；目标视角调整模块，被配置成基于上述距离信息将上述目标手部模型调整至目标视角。
以上描述仅为本公开的一些较佳实施例以及对所运用技术原理的说明。本领域技术人员应当理解，本公开的实施例中所涉及的发明范围，并不限于上述技术特征的特定组合而成的技术方案，同时也应涵盖在不脱离上述发明构思的情况下，由上述技术特征或其等同特征进行任意组合而形 成的其它技术方案。例如上述特征与本公开的实施例中公开的(但不限于)具有类似功能的技术特征进行互相替换而形成的技术方案。

Claims (11)

  1. 一种虚拟佩戴物匹配方法，包括：
    获取待处理图像和佩戴物图像，所述待处理图像包括手部图像，所述佩戴物图像包括佩戴物参数；
    对所述待处理图像进行识别，获取所述手部图像的关键点坐标；
    基于所述关键点坐标调整初始手部模型的参数，得到对应所述手部图像的目标手部模型；
    通过所述佩戴物参数构建虚拟佩戴物；
    将所述虚拟佩戴物与所述目标手部模型进行匹配，生成所述虚拟佩戴物的匹配图像。
  2. 根据权利要求1所述的方法，其中，所述待处理图像包括尺寸信息，所述尺寸信息用于指示所述手部图像对应的实际手部的尺寸，以及
    所述对所述待处理图像进行识别，获取所述手部图像的关键点坐标，包括：
    在所述手部图像上设置对应设定位置的至少一个指定标记点；
    基于所述尺寸信息和所述至少一个指定标记点确定所述手部图像的关键点坐标。
  3. 根据权利要求1所述的方法，其中，所述待处理图像包括点云数据，所述点云数据用于指示所述手部图像对应的手部在空间中的三维坐标，以及
    所述对所述待处理图像进行识别，获取所述手部图像的关键点坐标，包括：
    查找所述点云数据中对应设定位置的目标点云数据，并将所述目标点云数据的坐标标记为关键点坐标。
  4. 根据权利要求2或3所述的方法，其中，所述基于所述关键点坐 标调整初始手部模型的参数，得到对应所述手部图像的目标手部模型，包括：
    查找所述初始手部模型上对应所述设定位置的三维关键点；
    通过所述关键点坐标对所述三维关键点的坐标值进行调整，得到对应所述手部图像的目标手部模型。
  5. 根据权利要求1所述的方法，其中，所述佩戴物参数包括内径周长，以及
    所述将所述虚拟佩戴物与所述目标手部模型进行匹配，包括：
    确定所述目标手部空间上对应所述虚拟佩戴物的佩戴位置，并将所述虚拟佩戴物与所述佩戴位置对应设置；
    响应于所述佩戴位置的外径周长与所述内径周长的差值小于设置阈值，标记所述虚拟佩戴物与所述目标手部模型匹配。
  6. 根据权利要求5所述的方法，其中，所述将所述虚拟佩戴物与所述目标手部模型进行匹配，包括：
    响应于所述佩戴位置的外径周长与所述内径周长的差值大于等于设置阈值，调整所述佩戴物参数。
  7. 根据权利要求1所述的方法，所述生成所述虚拟佩戴物的匹配图像，包括：
    将设置所述虚拟佩戴物后的所述目标手部模型调整至目标视角，所述目标视角与所述待处理图像中手部图像的视角相同；
    生成对应所述目标视角的初始二维图像；
    将所述初始二维图像中的所述佩戴物图像添加至所述待处理图像，得到匹配图像。
  8. 根据权利要求7所述的方法，所述将设置所述虚拟佩戴物后的所述目标手部模型调整至目标视角，包括：
    从所述待处理图像中选择设定数量的关键点坐标作为目标关键点；
    测量所述目标关键点在所述待处理图像上的距离信息；
    基于所述距离信息将所述目标手部模型调整至目标视角。
  9. 一种虚拟佩戴物匹配装置，包括：
    图像获取单元，被配置成获取待处理图像和佩戴物图像，所述待处理图像包括手部图像，所述佩戴物图像包括佩戴物参数；
    关键点坐标获取单元，被配置成对所述待处理图像进行识别，获取所述手部图像的关键点坐标；
    目标手部模型获取单元，被配置成基于所述关键点坐标调整初始手部模型的参数，得到对应所述手部图像的目标手部模型；
    虚拟佩戴物构建单元，被配置成通过所述佩戴物参数构建虚拟佩戴物；
    匹配图像生成单元，被配置成将所述虚拟佩戴物与所述目标手部模型进行匹配，生成所述虚拟佩戴物的匹配图像。
  10. 一种电子设备，包括：
    一个或多个处理器；
    存储装置，其上存储有一个或多个程序，
    当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现如权利要求1至8中任一所述的方法。
  11. 一种计算机可读介质，其上存储有计算机程序，其中，所述程序被处理器执行时实现如权利要求1至8中任一所述的方法。
PCT/CN2021/109551 2020-10-14 2021-07-30 虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质 WO2022078014A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011095497.1A CN112232183B (zh) 2020-10-14 2020-10-14 虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质
CN202011095497.1 2020-10-14

Publications (1)

Publication Number Publication Date
WO2022078014A1 true WO2022078014A1 (zh) 2022-04-21

Family

ID=74112759

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/109551 WO2022078014A1 (zh) 2020-10-14 2021-07-30 虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质

Country Status (2)

Country Link
CN (1) CN112232183B (zh)
WO (1) WO2022078014A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112232183B (zh) * 2020-10-14 2023-04-28 抖音视界有限公司 虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质
CN112241999A (zh) * 2020-10-21 2021-01-19 北京字跳网络技术有限公司 图像生成方法、装置、设备和计算机可读介质
CN114067086A (zh) * 2021-10-22 2022-02-18 深圳市商汤科技有限公司 一种图像处理方法、装置及其相关设备和存储介质
CN115937964B (zh) * 2022-06-27 2023-12-15 北京字跳网络技术有限公司 姿态估计的方法、装置、设备和存储介质
CN115171097B (zh) * 2022-09-05 2022-12-09 中科航迈数控软件(深圳)有限公司 基于三维点云的加工控制方法、系统及相关设备
CN118521747A (zh) * 2023-02-17 2024-08-20 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

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102156808A (zh) * 2011-03-30 2011-08-17 北京触角科技有限公司 增强现实实时虚拟饰品试戴系统及方法
CN104820498A (zh) * 2015-05-14 2015-08-05 周谆 手部虚拟饰品试戴的人机交互方法和系统
CN107015659A (zh) * 2017-05-03 2017-08-04 湖南拓视觉信息技术有限公司 一种手表虚拟试戴方法和系统
US20170323374A1 (en) * 2016-05-06 2017-11-09 Seok Hyun Park Augmented reality image analysis methods for the virtual fashion items worn
CN110163048A (zh) * 2018-07-10 2019-08-23 腾讯科技（深圳）有限公司 手部关键点的识别模型训练方法、识别方法及设备
CN110210501A (zh) * 2019-06-11 2019-09-06 北京字节跳动网络技术有限公司 虚拟对象生成方法、电子设备及计算机可读存储介质
CN110221690A (zh) * 2019-05-13 2019-09-10 Oppo广东移动通信有限公司 基于ar场景的手势交互方法及装置、存储介质、通信终端
US20200184668A1 (en) * 2018-12-05 2020-06-11 Qualcomm Incorporated Systems and methods for three-dimensional pose determination
CN112232183A (zh) * 2020-10-14 2021-01-15 北京字节跳动网络技术有限公司 虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20190043362A (ko) * 2017-10-18 2019-04-26 고등기술연구원연구조합 맞춤형 가상현실공간 제공 시스템 및 방법
CN110111418B (zh) * 2019-05-15 2022-02-25 北京市商汤科技开发有限公司 创建脸部模型的方法、装置及电子设备
CN110363867B (zh) * 2019-07-16 2022-11-29 芋头科技（杭州）有限公司 虚拟装扮系统、方法、设备及介质
CN111325851B (zh) * 2020-02-28 2023-05-05 腾讯科技（深圳）有限公司 图像处理方法及装置、电子设备和计算机可读存储介质
CN111461814B (zh) * 2020-03-11 2024-03-26 深圳市新镜介网络有限公司 虚拟眼镜的试戴方法、终端设备及存储介质

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102156808A (zh) * 2011-03-30 2011-08-17 北京触角科技有限公司 增强现实实时虚拟饰品试戴系统及方法
CN104820498A (zh) * 2015-05-14 2015-08-05 周谆 手部虚拟饰品试戴的人机交互方法和系统
US20170323374A1 (en) * 2016-05-06 2017-11-09 Seok Hyun Park Augmented reality image analysis methods for the virtual fashion items worn
CN107015659A (zh) * 2017-05-03 2017-08-04 湖南拓视觉信息技术有限公司 一种手表虚拟试戴方法和系统
CN110163048A (zh) * 2018-07-10 2019-08-23 腾讯科技（深圳）有限公司 手部关键点的识别模型训练方法、识别方法及设备
US20200184668A1 (en) * 2018-12-05 2020-06-11 Qualcomm Incorporated Systems and methods for three-dimensional pose determination
CN110221690A (zh) * 2019-05-13 2019-09-10 Oppo广东移动通信有限公司 基于ar场景的手势交互方法及装置、存储介质、通信终端
CN110210501A (zh) * 2019-06-11 2019-09-06 北京字节跳动网络技术有限公司 虚拟对象生成方法、电子设备及计算机可读存储介质
CN112232183A (zh) * 2020-10-14 2021-01-15 北京字节跳动网络技术有限公司 虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质

Also Published As

Publication number Publication date
CN112232183A (zh) 2021-01-15
CN112232183B (zh) 2023-04-28

Similar Documents

Publication Publication Date Title
WO2022078014A1 (zh) 虚拟佩戴物匹配方法、装置、电子设备和计算机可读介质
US20230343055A1 (en) Reconstruction of 3d model with immersive experience
US8811951B1 (en) Managing display of private information
EP2891954B1 (en) User-directed personal information assistant
US11132842B2 (en) Method and system for synchronizing a plurality of augmented reality devices to a virtual reality device
JP2017513130A (ja) ウエアラブルデバイスの着用状態の処理方法及び装置
US20210279969A1 (en) Crowd sourced mapping system
TWI718631B (zh) 人臉圖像的處理方法及裝置、電子設備和儲存介質
CN111026314A (zh) 控制显示设备的方法及便携设备
CN110837607A (zh) 兴趣点匹配方法、装置、计算机设备以及存储介质
CN109683710B (zh) 一种手掌法向量确定方法、装置、设备及存储介质
US10139925B2 (en) Causing specific location of an object provided to a device
CN112204621A (zh) 基于计算设备能力简档的虚拟骨架
US10095726B2 (en) Multiple workspace database engine
CN110248197B (zh) 语音增强方法及装置
WO2023051244A1 (zh) 图像生成方法、装置、设备及存储介质
WO2023151558A1 (zh) 用于显示图像的方法、装置和电子设备
CN114092655A (zh) 构建地图的方法、装置、设备及存储介质
US20140149950A1 (en) Image overlay-based user interface apparatus and method
US20230230325A1 (en) Applying stored digital makeup enhancements to recognized faces in digital images
WO2024094158A1 (zh) 特效处理方法、装置、设备及存储介质
WO2022083213A1 (zh) 图像生成方法、装置、设备和计算机可读介质
US20240265641A1 (en) Augmented reality device for obtaining position information of joints of user's hand and operating method thereof
CN112183657B (zh) 获取标注信息的方法、装置、电子设备和计算机可读介质
US10559028B2 (en) Method, medium, and system for social network initiated listings

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: 21879061

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21879061

Country of ref document: EP

Kind code of ref document: A1