WO2023001039A1 - 图像匹配方法、装置、设备及存储介质 - Google Patents

图像匹配方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2023001039A1
WO2023001039A1 PCT/CN2022/105448 CN2022105448W WO2023001039A1 WO 2023001039 A1 WO2023001039 A1 WO 2023001039A1 CN 2022105448 W CN2022105448 W CN 2022105448W WO 2023001039 A1 WO2023001039 A1 WO 2023001039A1
Authority
WO
WIPO (PCT)
Prior art keywords
matched
image
frame
information
matching result
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/CN2022/105448
Other languages
English (en)
French (fr)
Inventor
吕绍辉
陈怡�
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.)
Beijing Zitiao Network Technology Co Ltd
Original Assignee
Beijing Zitiao Network Technology 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 Beijing Zitiao Network Technology Co Ltd filed Critical Beijing Zitiao Network Technology Co Ltd
Priority to US18/569,666 priority Critical patent/US20240282142A1/en
Publication of WO2023001039A1 publication Critical patent/WO2023001039A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • 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
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation
    • G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
    • 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
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172—Classification, e.g. identification
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • 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
    • G06V40/10—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 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
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1365—Matching; Classification
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Definitions

  • the present disclosure relates to the field of image processing, for example, to an image matching method, device, equipment and storage medium.
  • the present disclosure provides an image matching method, device, device, and storage medium, which can enrich the functions of image and video application program products and improve user experience.
  • the present disclosure provides an image matching method, the method comprising:
  • the matching result of the image to be matched is determined.
  • the present disclosure provides an image matching device, the device comprising:
  • the first identification module is configured to determine the image to be matched, and identify the target object on the image to be matched;
  • the first acquisition module is configured to, in response to recognizing that the target object exists on the image to be matched, determine a frame to be matched that contains the target object on the image to be matched, and obtain the corresponding information to be matched;
  • the first matching module is configured to match the to-be-matched information corresponding to the to-be-matched frame with the calibration frame information corresponding to the to-be-matched image to obtain a matching result of the calibration frame information; wherein the calibration frame information is based on The reference line displayed on the standard image corresponding to the image to be matched is calibrated to obtain;
  • the first determination module is configured to determine the matching result of the image to be matched based on the matching result of the calibration frame information.
  • the present disclosure provides a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and when the instructions are run on a terminal device, the terminal device is made to implement the above method.
  • the present disclosure provides a device, including: a memory, a processor, and a computer program stored on the memory and operable on the processor, when the processor executes the computer program, Implement the above method.
  • the present disclosure provides a computer program product, where the computer program product includes a computer program/instruction, and when the computer program/instruction is executed by a processor, the processor implements the above method.
  • FIG. 1 is a flowchart of an image matching method provided by an embodiment of the present disclosure
  • FIG. 2 is a schematic diagram showing a standard image with reference lines provided by an embodiment of the present disclosure
  • FIG. 3 is a schematic structural diagram of an image matching device provided by an embodiment of the present disclosure.
  • Fig. 4 is a schematic structural diagram of an image matching device provided by an embodiment of the present disclosure.
  • the embodiment of the present disclosure provides an image matching method.
  • the image to be matched is determined, and the target object is identified on the image to be matched. If it is determined that there is For the target object, a frame to be matched containing the target object is determined on the image to be matched, and information to be matched corresponding to the frame to be matched is acquired. Then, match the information to be matched corresponding to the frame to be matched with the information of the calibration frame corresponding to the image to be matched to obtain a matching result of the calibration frame information, and finally determine the matching result of the image to be matched based on the matching result.
  • the embodiment of the present disclosure determines the matching result of the image to be matched by matching the calibration frame information corresponding to the pre-marked image to be matched. Based on the image matching method provided by the embodiment of the present disclosure, the functions of image and video application products can be enriched. Improve user experience.
  • an embodiment of the present disclosure provides an image matching method.
  • FIG. 1 it is a flow chart of an image matching method provided by an embodiment of the present disclosure. The method includes:
  • S101 Determine an image to be matched, and identify a target object on the image to be matched.
  • the image to be matched in the embodiment of the present disclosure may be a frame of image obtained from continuous frame images captured by a camera, or may be an image uploaded by a user or received from another terminal.
  • the images to be matched may include portrait images, object images, and the like.
  • the target object in the embodiments of the present disclosure may include body parts such as human faces, hands, legs, etc., or may also be various items, or may also be preset text and the like.
  • the target object may include one or more.
  • the target object may include a human face, and it is sufficient to perform face recognition on the image to be matched.
  • face recognition and hand recognition can be performed on the image to be matched, and then the frame to be matched containing the face and the frame to be matched containing the hand can be obtained.
  • the image to be matched can be identified based on the characteristics of the target object, so as to determine whether there is a preset object on the image to be matched. For example, face recognition is performed on the image to be matched based on the features of the face to determine whether there is a face on the image to be matched.
  • a specific face recognition algorithm is not limited in this embodiment of the present disclosure.
  • a frame to be matched containing the target object is determined on the image to be matched.
  • the target object is a human face
  • a frame to be matched containing a human face is determined from the image to be matched, that is, a face frame to be matched.
  • the frame to be matched may be the smallest rectangular frame containing the target object on the image to be matched.
  • one or more frames to be matched may be determined on the image to be matched. For example, if multiple human faces are recognized on the image to be matched, multiple human face frames to be matched may be determined from the image to be matched, and each face frame to be matched contains a human face. For another example, if a human face and hands are recognized on the image to be matched, the frame to be matched containing the face and the frame to be matched containing the hand can be determined from the image to be matched.
  • the to-be-matched information corresponding to the frame to be matched is acquired.
  • the to-be-matched information corresponding to the to-be-matched frame may include to-be-matched position information of the to-be-matched frame, to-be-matched key point information, and other information that can be used to describe the characteristics of the target object in the to-be-matched frame.
  • the to-be-matched information corresponding to the to-be-matched face frame may include position information of the to-be-matched face frame, facial features key point information on the face in the to-be-matched face frame, and the like.
  • the to-be-matched information corresponding to the to-be-matched frame may be acquired through image recognition.
  • S103 Match the to-be-matched information corresponding to the to-be-matched frame with the marking frame information corresponding to the to-be-matched image, to obtain a matching result of the marking frame information.
  • the calibration frame information is obtained based on the reference line calibration displayed on the standard image corresponding to the image to be matched.
  • the information to be matched corresponding to the frame to be matched on the image to be matched is acquired, the information to be matched corresponding to the frame to be matched is matched Whether the calibration frame information corresponding to the matching image is successfully matched.
  • the marking frame information corresponding to the image to be matched is obtained based on the reference line calibration displayed on the standard image corresponding to the image to be matched.
  • the standard image corresponding to the image to be matched includes one or more reference lines.
  • the reference line on the standard image may be a posture line that can represent a portrait posture on the standard image.
  • FIG. 2 it is a schematic diagram showing a standard image with reference lines provided by an embodiment of the present disclosure.
  • the user can mark a calibration frame on the standard image, such as the calibration frame containing a human face as shown in FIG. 2 .
  • the calibration position information corresponding to the calibration frame is obtained, which is used to form the calibration frame information corresponding to the calibration frame.
  • the marking frame information corresponding to the standard image includes marking frame information respectively corresponding to the multiple marking frames.
  • the calibration position information is obtained from the calibration frame information corresponding to the image to be matched, and then the position information to be matched of the frame to be matched is matched with the calibration position information , obtaining the position matching result of the calibration frame corresponding to the calibration position information, and determining the matching result of the calibration frame information corresponding to the image to be matched based on the position matching result of the calibration frame. For example, when it is determined that the matching position information of the frame to be matched is successfully matched with the marked position information, it is determined that the marked frame corresponding to the marked position information is successfully matched.
  • the calibration frame information may also include calibration key point information pre-configured for the calibration frame, which is used to characterize the characteristics of the target object in the calibration frame.
  • the target object can be a human face, hand, foot, etc.
  • the information may include face key point information, hand key point information, or foot key point information.
  • the calibration frame is a calibration frame for the face
  • the calibration key point information configured in advance for the calibration frame can be the key point information of facial features, which is used to represent the face in the calibration frame that needs to be matched.
  • expressive features For example, the key point information of facial features may include position information of key points of facial features and information about their positional relationship, etc., which can represent feature information of facial expressions.
  • the pre-configured calibration key point information for the calibration frame includes calibration action key point information, which is used to represent the hand or foot that needs to be matched in the calibration frame. action features.
  • the to-be-matched position information of the to-be-matched frame is matched with the marked position information to obtain a position matching result of the marked frame corresponding to the marked position information. If it is determined that the position matching result is a successful match, then matching the key point information to be matched in the frame to be matched with the key point information corresponding to the calibration frame to obtain the key point matching result of the calibration frame, And based on the key point matching result of the calibration frame, determine the matching result of the calibration frame information corresponding to the image to be matched.
  • the marking position information and marking key point information corresponding to the marking frame may be used to form the marking frame information corresponding to the marking frame.
  • the calibration box information is saved in the form of a json file. The present disclosure does not limit the storage form of the calibration frame information.
  • the frame to be matched on the image to be matched includes a human face
  • the information to be matched is matched with the information of the calibration frame corresponding to the image to be matched
  • the to-be-matched information corresponding to the to-be-matched frame may include face key point information in the to-be-matched frame, position information of the to-be-matched frame, and the like.
  • the face key point information in the frame to be matched is extracted and the position information of the frame to be matched is obtained
  • the face key point information and position information are respectively matched with the calibration frame information corresponding to the image to be matched
  • the calibration frame information includes calibration key point information and calibration position information. If it is determined that the face key point information and position information of the frame to be matched are successfully matched with the calibration frame information of the same calibration frame, it means that the calibration frame is successfully matched. When it is determined that at least one of the calibration key point information and the calibration position information in the calibration frame information of the calibration frame does not match successfully, it may be determined that the matching result of the calibration frame is a matching failure.
  • the algorithm for extracting the key points of the human face is not limited in the embodiments of the present disclosure.
  • the image to be matched may include a plurality of frames to be matched, and for each frame to be matched, the corresponding information to be matched is determined respectively, and the information to be matched corresponding to each frame to be matched is combined with the frame to be matched Match the marking frame information corresponding to the image to determine the matching result of the marking frame information corresponding to the image to be matched.
  • S104 Based on the matching result of the marking frame information, determine the matching result of the image to be matched.
  • the matching result of the marking frame information is a matching failure. Therefore, it may also be determined that the matching result of the image to be matched is also a matching failure.
  • the image to be matched is determined, and the target object is identified on the image to be matched, and if it is determined that the target object exists on the image to be matched, then it is determined that the target is included in the image to be matched Object's to-be-matched box, and obtain the to-be-matched information corresponding to the to-be-matched box. Then, match the matching information corresponding to the matching frame with the calibration frame information corresponding to the image to be matched to obtain the matching result of the calibration frame information, and finally, determine the matching result of the image to be matching based on the matching result.
  • the embodiment of the present disclosure determines the matching result of the image to be matched by matching the calibration frame information corresponding to the pre-marked image to be matched. Based on the image matching method provided by the embodiment of the present disclosure, the functions of image and video application products can be enriched. Improve user experience.
  • the marking frame information corresponding to the image to be matched may include marking frame information corresponding to multiple marking frames respectively. Take, for example, a standard image showing guide lines corresponding to multiple objects of interest. For example, based on the reference line displayed on the standard image, the calibration frame information of the calibration frames corresponding to the three faces can be calibrated, and the matching information of the frame to be matched on the image to be matched is combined with the pre-calibrated 3 faces. The calibration frame information of the calibration frames corresponding to the faces are matched, and the matching result of the calibration frame information corresponding to the image to be matched can be obtained.
  • the calibration frame corresponding to the image to be matched can be determined
  • the matching result of the information is a successful matching, and it is further determined that the matching result of the image to be matched is a successful matching.
  • the marking frame information corresponding to the image to be matched includes marking frame information corresponding to multiple marking frames respectively.
  • the information to be matched is matched with the frame information of the frame to be matched that is not marked by the preset mark in the plurality of frames, and when the matching is successful , mark the preset identifier for the calibration frame that is successfully matched with the information to be matched corresponding to the frame to be matched, to indicate that the calibration frame has been matched successfully.
  • the matching result of the marking frame information is determined.
  • the matching result of the marked frame information is a successful match; if it is determined that at least one of the multiple marked frames is not marked with The preset identification determines that the matching result of the calibration frame information is a matching failure.
  • the image matching method provided by the embodiments of the present disclosure matches the calibration frame information corresponding to the multiple calibration frames corresponding to the pre-calibrated image to be matched, which can realize multi-person interaction, enrich the functions of image and video application products, and improve User experience.
  • the embodiment of the present disclosure also provides an embodiment of an application scenario of an image matching method, wherein the image to be matched in the embodiment of the present disclosure may belong to one frame of consecutive frame images within a preset first time period
  • a preset first time period For an image, for example, 0-1 second when the camera is turned on and starts shooting is the preset first time period, and 1-2 seconds can also be used as the preset first time period.
  • the matching result of each frame image in the continuous frame images within the preset first time period can be determined, and then, based on the matching result of each frame image, finally determine The matching result image corresponding to the preset first time period is displayed.
  • the image to be matched is determined as the matching result image corresponding to the preset first time period. If it is determined that the matching result of the image to be matched within the preset first time period is a matching failure, updating the image to be matched based on the next frame image of the image to be matched within the preset first time period, Continue triggering and executing the step of identifying the target object on the image to be matched until it is determined that the matching result of the image to be matched is successful.
  • the 8th frame image can be determined as the matching result corresponding to the time period of 0-1 second image, and no longer match against subsequent frame images within 0-1 seconds. If the matching result of the 8th frame image is a matching failure, continue to match the 9th frame image in the time period of 0-1 second until the first image that is successfully matched in the time period of 0-1 second is obtained, And determine it as the matching result image corresponding to the time period of 0-1 second.
  • the preset first time period belongs to one of a plurality of preset time periods, assuming that the plurality of preset time periods include a preset first time period, a preset second time period and a preset third time period, and the preset first time period, preset second time period and preset third time period are three consecutive time periods, for example, 0-1 seconds, 1-2 seconds and 2 seconds when the camera is turned on to start shooting -3 seconds.
  • a preset number of target output images are determined from each of the determined matching result images, and then the preset number of target output images are displayed . For example, from the matching result images respectively corresponding to the preset first time period, the preset second time period and the preset third time period, two target output images are determined for displaying on the device interface.
  • a matching result image whose matching result is a successful matching is preferentially displayed on the device interface.
  • a preset number of matching result images that are successfully matched may be randomly determined from the matching result images corresponding to a plurality of preset time periods, and displayed on the device interface.
  • a preset number of slots are predetermined for storing target output images, that is, images displayed on the device interface.
  • the slot can be understood as a hole in the storage space, which is used to store an object or object. When other objects or objects enter, the original object or object is squeezed out.
  • Each slot in the embodiment of the present disclosure is used to store an image, assuming that image A has been stored in the slot, if image B is stored in the slot, the image A in the slot will be squeezed out, at this time This slot is occupied by image B.
  • the matching result image After determining the matching result image corresponding to the preset first time period, it is first determined whether there is a free slot in the preset number of slots, and if there is a free slot, the matching result image is stored in the free slot. If there is no free slot, the occupied slot is updated based on the matching result of the matching result image.
  • the occupied slot is updated, and a preset number of images can be randomly determined from the matching result image and the image occupying the slot, and The determined preset number of images is stored in the slot.
  • the image matching method provided by the embodiment of the present disclosure displays the image occupying the slot on the device interface, and can achieve the effect of preferentially and randomly displaying the matched image on the device interface.
  • the present disclosure also provides an image matching device.
  • FIG. 3 it is a schematic structural diagram of an image matching device provided by an embodiment of the present disclosure.
  • the device includes:
  • the first recognition module 301 is configured to determine the image to be matched, and identify the target object on the image to be matched;
  • the first acquiring module 302 is configured to, in response to recognizing that the target object exists on the image to be matched, determine a frame to be matched containing the target object on the image to be matched, and acquire the frame to be matched Corresponding information to be matched;
  • the first matching module 303 is configured to match the to-be-matched information corresponding to the to-be-matched frame with the marking frame information corresponding to the to-be-matched image to obtain a matching result of the marking frame information; wherein, the marking frame information Obtained based on the reference line calibration displayed on the standard image corresponding to the image to be matched;
  • the first determining module 304 is configured to determine the matching result of the image to be matched based on the matching result of the marking frame information.
  • the information of the marking frame includes marking position information of the marking frame obtained based on the calibration of the reference line displayed on the standard image, and the information to be matched corresponding to the frame to be matched includes the information to be matched The location information of the frame to be matched;
  • the first matching module includes:
  • the first acquisition submodule is configured to acquire the calibration position information in the calibration frame information corresponding to the image to be matched;
  • the first matching submodule is configured to match the position information to be matched with the marked position information of the frame to be matched, and obtain the position matching result of the marked frame corresponding to the marked position information;
  • the first determining submodule is configured to determine a matching result of the marking frame information corresponding to the image to be matched based on the position matching result of the marking frame.
  • the calibration frame information further includes calibration key point information pre-configured for the calibration frame, and the to-be-matched information corresponding to the to-be-matched frame includes the to-be-matched key point information in the to-be-matched frame ;
  • the first matching module also includes:
  • the second matching submodule is configured to match the key point information to be matched in the frame to be matched with the calibrated key point information corresponding to the calibration frame when it is determined that the position matching result is a successful match, to obtain the The key point matching result of the calibration frame;
  • the first determining submodule is set to:
  • the matching result of the calibration frame information corresponding to the image to be matched is determined.
  • the target object includes hands and/or feet
  • the marked key point information includes hand key point information and/or foot key point information
  • the marking frame information corresponding to the image to be matched includes marking frame information corresponding to multiple marking frames respectively;
  • the first matching module includes:
  • the second determining submodule is configured to determine the marking frame information of the marking frame that is not marked with a preset identifier from the marking frame information corresponding to the plurality of marking frames respectively; wherein the preset identifier is used to represent the corresponding The calibration frame has been matched successfully;
  • the third matching submodule is configured to match the to-be-matched information corresponding to the to-be-matched frame based on the calibration frame information of the unmarked calibration frame, and when the matching is successful, match the to-be-matched information with the to-be-matched frame
  • the matching information corresponding to the matching frame is successfully matched with the calibration frame marking the preset identifier
  • the third determining submodule is configured to determine the matching result of the marking frame information based on the labeling of the multiple marking frames with respect to the preset identifier.
  • the third determining submodule is set to:
  • the matching result of the marked frame information is a successful match; if it is determined that at least one of the multiple marked frames is not marked with the A preset flag is used to determine that the matching result of the calibration box information is a matching failure.
  • the image to be matched belongs to one frame image in the continuous frame images within a preset first time period; the device further includes:
  • the second determination module is configured to determine the image to be matched as the matching result image corresponding to the preset first time period when it is determined that the matching result of the image to be matched is a successful match;
  • a first update module configured to update the image to be matched based on the next frame image of the image to be matched within the preset first time period when the matching result of the image to be matched is determined to be a matching failure , continue to trigger the first identification module until it is determined that the matching result of the image to be matched is successful.
  • the preset first time period belongs to one of multiple preset time periods, and the device further includes:
  • the third determination module is configured to determine a preset number of target output images based on the matching result images respectively corresponding to the plurality of preset time periods;
  • the display module is configured to display the preset number of target output images.
  • the third determination module includes:
  • the fourth determining submodule is configured to determine whether there are free slots among the preset number of slots after determining the matching result image corresponding to the preset first time period; wherein, the preset number The slots are used to store the preset number of target output images;
  • the first storing submodule is configured to store the matching result image corresponding to the preset first time period into the free slot when it is determined that there are free slots in the preset number of slots;
  • the first update submodule is configured to update the preset number of slots based on the matching result of the matching result image when it is determined that there is no free slot in the preset number of slots.
  • the first updating submodule is configured to, when it is determined that there are no free slots in the preset number of slots, from the matching result image and the preset number of Among the images in the slots, the preset number of images is randomly determined, and the preset number of images are respectively stored in the preset number of slots.
  • the image to be matched is determined, and the target object is identified on the image to be matched, and if it is determined that the target object exists on the image to be matched, it is determined that the target is included in the image to be matched Object's to-be-matched box, and obtain the to-be-matched information corresponding to the to-be-matched box. Then, match the information to be matched corresponding to the frame to be matched with the information of the marking frame corresponding to the image to be matched to obtain a matching result of the marking frame information, and then determine the matching result of the image to be matched based on the matching result.
  • the embodiment of the present disclosure determines the matching result of the image to be matched by matching the calibration frame information corresponding to the pre-marked image to be matched, based on the image matching device provided by the embodiment of the present disclosure, it can enrich the functions of image and video application products, Improve user experience.
  • an embodiment of the present disclosure also provides a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and when the instructions are run on a terminal device, the terminal device realizes this The image matching method described in the embodiment is disclosed.
  • the computer readable storage medium may be a non-transitory computer readable storage medium.
  • the embodiments of the present disclosure also provide a computer program product, including computer programs/instructions, and when the computer programs/instructions are executed by a processor, the image matching method described in the embodiments of the present disclosure is implemented.
  • an embodiment of the present disclosure also provides an image matching device, as shown in FIG. 4 , which may include:
  • Processor 401 memory 402 , input device 403 and output device 404 .
  • the number of processors 401 in the image matching device can be one or more, and one processor is taken as an example in FIG. 4 .
  • the processor 401 , the memory 402 , the input device 403 and the output device 404 may be connected via a bus or in other ways, wherein connection via a bus is taken as an example in FIG. 4 .
  • the memory 402 can be configured to store software programs and modules, and the processor 401 executes various functional applications and data processing of the image matching device by running the software programs and modules stored in the memory 402 .
  • the memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function, and the like.
  • the memory 402 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.
  • the input device 403 can be used to receive input numbers or character information, and generate signal input related to user settings and function control of the image matching device.
  • the processor 401 loads the executable file corresponding to the process of one or more application programs into the memory 402 according to the following instructions, and the processor 401 runs the application stored in the memory 402 program, so as to realize the multiple functions of the above-mentioned image matching device.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Health & Medical Sciences (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Analysis (AREA)

Abstract

本公开提供了一种图像匹配方法、装置、设备及存储介质，所述方法包括：首先，确定待匹配图像，并对待匹配图像进行目标对象的识别，响应于确定待匹配图像上存在该目标对象，在待匹配图像上确定包含有该目标对象的待匹配框，并获取该待匹配框对应的待匹配信息。将待匹配框对应的待匹配信息与待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果，基于该匹配结果确定待匹配图像的匹配结果。

Description

图像匹配方法、装置、设备及存储介质
本申请要求在2021年7月19日提交中国专利局、申请号为202110812807.5的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本公开涉及图像处理领域，例如涉及一种图像匹配方法、装置、设备及存储介质。
背景技术
随着图像和视频处理技术的不断发展，图像视频类应用程序产品越来越多，支持的玩法也越来越丰富。
目前，如何进一步丰富图像视频类应用程序产品的功能，提升用户的使用体验，是亟需解决的技术问题。
发明内容
本公开提供了一种图像匹配方法、装置、设备及存储介质，能够丰富图像视频类应用程序产品的功能，提升用户的使用体验。
第一方面，本公开提供了一种图像匹配方法，所述方法包括：
确定待匹配图像，并对所述待匹配图像进行目标对象的识别；
响应于识别到所述待匹配图像上存在所述目标对象，在所述待匹配图像上确定包含有所述目标对象的待匹配框，并获取所述待匹配框对应的待匹配信息；
将所述待匹配框对应的待匹配信息与所述待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果；其中，所述标定框信息基于所述待匹配图像对应的标准图像上展示的参考线标定得到；
基于所述标定框信息的匹配结果，确定所述待匹配图像的匹配结果。
第二方面，本公开提供了一种图像匹配装置，所述装置包括：
第一识别模块，设置为确定待匹配图像，并对所述待匹配图像进行目标对象的识别；
第一获取模块，设置为响应于识别到所述待匹配图像上存在所述目标对象，在所述待匹配图像上确定包含有所述目标对象的待匹配框，并获取所述待匹配框对应的待匹配信息；
第一匹配模块，设置为将所述待匹配框对应的待匹配信息与所述待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果；其中，所述标定框信息基于所述待匹配图像对应的标准图像上展示的参考线标定得到；
第一确定模块，设置为基于所述标定框信息的匹配结果，确定所述待匹配图像的匹配结果。
第三方面，本公开提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当所述指令在终端设备上运行时，使得所述终端设备实现上述的方法。
第四方面，本公开提供了一种设备，包括：存储器，处理器，及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时，实现上述的方法。
第五方面，本公开提供了一种计算机程序产品，所述计算机程序产品包括计算机程序/指令，所述计算机程序/指令被处理器执行时使得所述处理器实现上述的方法。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本公开的实施例，并与说明书一起用于解释本公开的原理。
图1为本公开实施例提供的一种图像匹配方法的流程图；
图2为本公开实施例提供的一种展示有参考线的标准图像的示意图；
图3为本公开实施例提供的一种图像匹配装置的结构示意图；
图4为本公开实施例提供的一种图像匹配设备的结构示意图。
具体实施方式
下面将对本公开的方案进行进一步描述。需要说明的是，在不冲突的情况下，本公开的实施例及实施例中的特征可以相互组合。
在下面的描述中阐述了很多具体细节以便于充分理解本公开，但本公开还可以采用其他不同于在此描述的方式来实施；显然，说明书中的实施例只是本公开的一部分实施例，而不是全部的实施例。
为了不断丰富图像视频类应用程序产品所支持的玩法，本公开实施例提供了一种图像匹配方法，首先，确定待匹配图像，并对待匹配图像进行目标对象的识别，如果确定待匹配图像上存在该目标对象，则在待匹配图像上确定包含有该目标对象的待匹配框，并获取该待匹配框对应的待匹配信息。然后，将待匹配框对应的待匹配信息与待匹配图像对应的标定框信息进行匹配，得到标定框信息的匹配结果，最终，基于该匹配结果确定待匹配图像的匹配结果。本公开实施例通过与预先标定的待匹配图像对应的标定框信息进行匹配，确定待匹配图像的匹配结果，基于本公开实施例提供的图像匹配方法，能够丰富图像视频类应用程序产品的功能，提升用户的使用体验。
为此，本公开实施例提供了一种图像匹配方法，参考图1，为本公开实施例提供的一种图像匹配方法的流程图，该方法包括：
S101：确定待匹配图像，并对所述待匹配图像进行目标对象的识别。
本公开实施例中的待匹配图像可以为从摄像头拍摄到的连续帧图像中获取到的一帧图像，也可以为用户上传或者从其他终端接收到的图像等。例如，待匹配图像可以包括人像图像、物体图像等。
本公开实施例中的目标对象可以包括诸如人脸、手部、腿部等身体部位，或者也可以为多种物品，或者还可以为预设文字等。其中，目标对象可以包括一个或多个。例如，目标对象可以包括人脸，则对待匹配图像进行人脸识别即可。又例如目标对象可以同时包括人脸和手部，则对待匹配图像可以进行人脸识别和手部识别，后续可以得到包含人脸的待匹配框和包含手部的待匹配框。
实际应用中，可以基于目标对象的特征对待匹配图像进行识别，以确定待匹配图像上是否存在预设对象。例如，基于人脸的特征对待匹配图像进行人脸识别，以确定待匹配图像上是否存在人脸。具体的人脸识别算法在本公开实施例中不做限定。
S102：如果识别到所述待匹配图像上存在所述目标对象，则在所述待匹配图像上确定包含有所述目标对象的待匹配框，并获取所述待匹配框对应的待匹配信息。
本公开实施例中，在确定待匹配图像上存在目标对象时，在该待匹配图像上确定出包含有该目标对象的待匹配框。例如，假设目标对象为人脸，当在待匹配图像上识别到人脸时，则从待匹配图像上确定出包含有人脸的待匹配框，即待匹配人脸框。其中，待匹配框可以为待匹配图像上包含有目标对象的最小矩形框。
一种示例实施方式中，在待匹配图像上可以确定出一个或多个待匹配框。例如，在待匹配图像上识别到多个人脸，则可以从该待匹配图像上确定出多个待匹配人脸框，每个待匹配人脸框中分别包含有一张人脸。又例如，在待匹配图像上识别到人脸和手部，则可以从该待匹配图像上确定出包含人脸的待匹配框和包含手部的待匹配框。
本公开实施例中，在待匹配图像上确定出待匹配框之后，获取待匹配框对应的待匹配信息。其中，待匹配框对应的待匹配信息可以包括该待匹配框的待匹配位置信息、待匹配关键点信息等能够用于描述该待匹配框中的目标对象的特征的信息。
例如，待匹配人脸框对应的待匹配信息可以包括该待匹配人脸框的位置信息以及该待匹配人脸框中的人脸上的五官关键点信息等。例如，可以通过图像识别的方式等获取待匹配框对应的待匹配信息。
S103：将所述待匹配框对应的待匹配信息与所述待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果。
其中，所述标定框信息基于所述待匹配图像对应的标准图像上展示的参考线标定得到。
本公开实施例中，在获取到待匹配图像上的待匹配框对应的待匹配信息之后，将待匹配框对应的待匹配信息与该待匹配图像对应的标定框信息进行匹配，以确定该待匹配图像对应的标定框信息是否匹配成功。
一种示例实施方式中，待匹配图像对应的标定框信息基于该待匹配图像对应的标准图像上展示的参考线标定得到。通常，待匹配图像对应的标准图像包括一条或多条参考线，例如，标准图像上的参考线可以为能够表示出标准图像上的人像姿势的姿势线等。
如图2所示，为本公开实施例提供的一种展示有参考线的标准图像的示意图。基于标准图像上展示的参考线，用户可以在该标准图像上标定出标定框，如图2所示的包含有人脸的标定框。通过识别该标定框的位置坐标，得到该标定框对应的标定位置信息，用于构成该标定框对应的标定框信息。如果标准图像上标定出多个标定框，则该标准图像对应的标定框信息包括多个标定框分别对应的标定框信息。
实际应用中，在获取到待匹配图像对应的标定框信息之后，从该待匹配图像对应的标定框信息中获取标定位置信息，然后将待匹配框的待匹配位置信息与该标定位置信息进行匹配，得到该标定位置信息对应的标定框的位置匹配结果，并基于所述标定框的位置匹配结果，确定待匹配图像对应的标定框信息的匹配结果。例如，在确定待匹配框的待匹配位置信息与标定位置信息匹配成功时，则确定该标定位置信息对应的标定框匹配成功。
另外，标定框信息还可以包括预先为标定框配置的标定关键点信息，用于表征该标定框中的目标对象的特征，例如，目标对象可以为人脸、手部、脚部等，标定关键点信息可以包括人脸关键点信息、手部关键点信息或脚部关键点信息等。
如图2所示，标定框为针对人脸的标定框，则预先为该标定框配置的标定关键点信息可 以为人脸五官关键点信息，用于表征该标定框中的人脸需要匹配到的表情特征。例如，人脸五官关键点信息可以包括人脸五官关键点的位置信息以及相互之间的位置关系信息等，能够表征出人脸表情的特征信息。
假设标定框为针对手部或脚部的标定框，则预先为该标定框配置的标定关键点信息包括标定动作关键点信息，用于表征该标定框中需要匹配到的手部或脚部的动作特征。
本公开实施例中，首先将所述待匹配框的待匹配位置信息与所述标定位置信息进行匹配，得到所述标定位置信息对应的标定框的位置匹配结果。如果确定所述位置匹配结果为匹配成功，则将所述待匹配框中的待匹配关键点信息与所述标定框对应的标定关键点信息进行匹配，得到所述标定框的关键点匹配结果，并基于所述标定框的关键点匹配结果，确定所述待匹配图像对应的标定框信息的匹配结果。
实际应用中，将所述待匹配框的待匹配位置信息与标定位置信息进行匹配，得到该标定位置信息对应的标定框的位置匹配结果。然后，在确定位置匹配结果为匹配成功时，将待匹配关键点信息与该标定框对应的标定关键点信息进行匹配，得到该标定框的关键点匹配结果。如果确定位置匹配结果为匹配失败，则无需继续针对该待匹配框的待匹配关键点信息进行匹配，以此能够提高匹配效率，节省系统资源。
一种示例实施方式中，可以利用标定框对应的标定位置信息和标定关键点信息构成该标定框对应的标定框信息。通常，标定框信息是以json文件的形式保存的。本公开对标定框信息的保存形式不进行限制。
为了便于理解，假设待匹配图像上的待匹配框中包括人脸，则在获取到该待匹配框对应的待匹配信息后，将该待匹配信息与该待匹配图像对应的标定框信息进行匹配。其中，待匹配框对应的待匹配信息可以包括该待匹配框中的人脸关键点信息和该待匹配框的位置信息等。例如，在提取到待匹配框中的人脸关键点信息以及获取到该待匹配框的位置信息后，将人脸关键点信息和位置信息信息分别与该待匹配图像对应的标定框信息进行匹配，其中，标定框信息中包括标定关键点信息和标定位置信息。如果确定该待匹配框的人脸关键点信息和位置信息与同一标定框的标定框信息匹配成功，则说明该标定框匹配成功。在确定该标定框的标定框信息中的标定关键点信息和标定位置信息中至少存在一个未匹配成功时，则可以确定该标定框的匹配结果为匹配失败。其中，提取人脸关键点的算法在本公开实施例中不做限定。
一种示例实施方式中，待匹配图像上可以包括多个待匹配框，针对每个待匹配框，分别确定对应的待匹配信息，并将每个待匹配框对应的待匹配信息与该待匹配图像对应的标定框信息进行匹配，以确定该待匹配图像对应的标定框信息的匹配结果。
S104：基于所述标定框信息的匹配结果，确定所述待匹配图像的匹配结果。
本公开实施例中，在将待匹配图像上的待匹配框对应的待匹配信息均与该待匹配图像对应的标定框信息进行匹配之后，如果确定该待匹配图像对应的标定框信息匹配成功，则可以确定该待匹配图像的匹配结果为匹配成功。
如果确定该待匹配图像对应的标定框信息中存在未匹配成功的标定框信息，则确定标定框信息的匹配结果为匹配失败，因此，也可以确定该待匹配图像的匹配结果也为匹配失败。
本公开实施例提供的图像匹配方法中，首先，确定待匹配图像，并对待匹配图像进行目标对象的识别，如果确定待匹配图像上存在该目标对象，则在待匹配图像上确定包含有该目标对象的待匹配框，并获取该待匹配框对应的待匹配信息。然后，将待匹配框对应的待匹配 信息与待匹配图像对应的标定框信息进行匹配，得到标定框信息的匹配结果，最终，基于该匹配结果确定待匹配图像的匹配结果。本公开实施例通过与预先标定的待匹配图像对应的标定框信息进行匹配，确定待匹配图像的匹配结果，基于本公开实施例提供的图像匹配方法，能够丰富图像视频类应用程序产品的功能，提升用户的使用体验。
一种示例实施方式中，待匹配图像对应的标定框信息可以包括多个标定框分别对应的标定框信息。以展示有多个目标对象对应的参考线的标准图像为例。例如，基于该标准图像上展示的参考线可以标定得到3张人脸分别对应的标定框的标定框信息，通过将待匹配图像上的待匹配框的待匹配信息与预先标定得到的3张人脸分别对应的标定框的标定框信息进行匹配，可以得到该待匹配图像对应的标定框信息的匹配结果。
假设待匹配图像上包括3张人脸，且3张人脸的待匹配信息分别与预先标定的3张人脸的标定框的标定框信息匹配成功，则可以确定该待匹配图像对应的标定框信息的匹配结果为匹配成功，进一步的确定该待匹配图像的匹配结果为匹配成功。
一种示例实施方式中，待匹配图像对应的标定框信息中包括多个标定框分别对应的标定框信息。在获取到待匹配图像上的待匹配框对应的待匹配信息之后，将待匹配信息与该多个标定框中未被标记预设标识的标定框的标定框信息进行匹配，并在匹配成功时，为与该待匹配框对应的待匹配信息匹配成功的标定框标记该预设标识，以表明该标定框已匹配成功。基于待匹配图像对应的每个标定框针对该预设标识的标记情况，确定标定框信息的匹配结果。
例如，如果确定所述多个标定框均标记有所述预设标识，则确定所述标定框信息的匹配结果为匹配成功；如果确定所述多个标定框中的至少一个标定框未标记有所述预设标识，确定所述标定框信息的匹配结果为匹配失败。
本公开实施例提供的图像匹配方法通过与预先标定的待匹配图像对应的多个标定框分别对应的标定框信息进行匹配，能够实现多人交互功能，丰富图像视频类应用程序产品的功能，提升用户的使用体验。
基于上述实施例，本公开实施例还提供了一种图像匹配方法的应用场景实施例，其中，本公开实施例中的待匹配图像可以属于预设第一时间段内连续帧图像中的一帧图像，例如，在打开摄像头开始拍摄的0-1秒为预设第一时间段，1-2秒也可以作为预设第一时间段。
实际应用中，基于本公开实施例提供的图像匹配方法能够确定出预设第一时间段内连续帧图像中的每一帧图像的匹配结果，然后，基于每一帧图像的匹配结果，最终确定出预设第一时间段对应的匹配结果图像。
一种示例实施方式中，如果确定预设第一时间段内的待匹配图像的匹配结果为匹配成功，则将该待匹配图像确定为该预设第一时间段对应的匹配结果图像。如果确定预设第一时间段内的待匹配图像的匹配结果为匹配失败，则基于在所述预设第一时间段内的所述待匹配图像的下一帧图像更新所述待匹配图像，继续触发执行所述对所述待匹配图像进行目标对象的识别的步骤，直到确定所述待匹配图像的匹配结果为匹配成功。
假设待匹配图像为0-1秒内的第8帧图像，如果该第8帧图像的匹配结果为匹配成功，则可以将该第8帧图像确定为0-1秒的时间段对应的匹配结果图像，不再针对0-1秒内的后续帧图像进行匹配。如果该第8帧图像的匹配结果为匹配失败，则继续对0-1秒的时间段内的第9帧图像进行匹配，直到获取到0-1秒的时间段内首个匹配成功的图像，并将其确定为0-1秒时间段对应的匹配结果图像。
本公开实施例中，预设第一时间段属于多个预设时间段中的一个，假设多个预设时间段包括预设第一时间段、预设第二时间段和预设第三时间段，且预设第一时间段、预设第二时间段和预设第三时间段为连续的3个时间段，例如分别为打开摄像头开始拍摄的0-1秒、1-2秒和2-3秒。
实际应用中，在确定每个预设时间段分别对应的匹配结果图像后，从确定的各个匹配结果图像中确定出预设个数的目标输出图像，然后，显示预设个数的目标输出图像。例如，从预设第一时间段、预设第二时间段和预设第三时间段分别对应的匹配结果图像中，确定2张目标输出图像，用于显示于设备界面上。
一种示例实施方式中，为了提高用户体验，优先在设备界面上显示匹配结果为匹配成功的匹配结果图像。例如，可以从多个预设时间段分别对应的匹配结果图像中，随机确定预设个数的匹配成功的匹配结果图像，显示于设备界面上。
另一种示例实施方式中，预先确定预设个数的槽位，用于存储目标输出图像，即显示在设备界面上的图像。其中，槽位可以理解为存储空间中的一个坑位，用于存放一个物体或对象，当其他物体或对象进入后，原有物体或对象则被挤出。本公开实施例中的每个槽位用于存储一张图像，假设槽位中已存储图像A，如果图像B存入该槽位，则该槽位中的图像A会被挤出，此时该槽位被图像B占用。在确定预设第一时间段对应的匹配结果图像之后，首先确定预设个数的槽位中是否存在空闲槽位，如果存在空闲槽位，则将该匹配结果图像存入空闲槽位中。如果不存在空闲槽位，则基于该匹配结果图像的匹配结果，更新被占用的槽位。
在一种示例实施例中，基于该匹配结果图像的匹配结果，更新被占用的槽位，可以从该匹配结果图像和占用槽位的图像中，随机确定出预设个数的图像，并将确定出的预设个数的图像存储于槽位中。
本公开实施例提供的图像匹配方法将占用槽位的图像显示于设备界面上，能够实现优先且随机的在设备界面上显示匹配成功的图像的效果。
基于上述方法实施例，本公开还提供了一种图像匹配装置，参考图3，为本公开实施例提供的一种图像匹配装置的结构示意图，所述装置包括：
第一识别模块301，设置为确定待匹配图像，并对所述待匹配图像进行目标对象的识别；
第一获取模块302，设置为响应于识别到所述待匹配图像上存在所述目标对象，在所述待匹配图像上确定包含有所述目标对象的待匹配框，并获取所述待匹配框对应的待匹配信息；
第一匹配模块303，设置为将所述待匹配框对应的待匹配信息与所述待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果；其中，所述标定框信息基于所述待匹配图像对应的标准图像上展示的参考线标定得到；
第一确定模块304，设置为基于所述标定框信息的匹配结果，确定所述待匹配图像的匹配结果。
一种示例实施方式中，所述标定框信息包括基于所述标准图像上展示的所述参考线标定得到的标定框的标定位置信息，所述待匹配框对应的待匹配信息包括所述待匹配框的待匹配位置信息；
所述第一匹配模块，包括：
第一获取子模块，设置为获取所述待匹配图像对应的标定框信息中的所述标定位置信息；
第一匹配子模块，设置为将所述待匹配框的待匹配位置信息与所述标定位置信息进行匹 配，得到所述标定位置信息对应的标定框的位置匹配结果；
第一确定子模块，设置为基于所述标定框的位置匹配结果，确定所述待匹配图像对应的标定框信息的匹配结果。
一种示例实施方式中，所述标定框信息还包括预先为所述标定框配置的标定关键点信息，所述待匹配框对应的待匹配信息包括所述待匹配框中的待匹配关键点信息；
所述第一匹配模块，还包括：
第二匹配子模块，设置为在确定所述位置匹配结果为匹配成功时，将所述待匹配框中的待匹配关键点信息与所述标定框对应的标定关键点信息进行匹配，得到所述标定框的关键点匹配结果；
所述第一确定子模块，设置为：
基于所述标定框的关键点匹配结果，确定所述待匹配图像对应的标定框信息的匹配结果。
一种示例实施方式中，所述目标对象包括手部和/或脚部，所述标定关键点信息包括手部关键点信息和/或脚部关键点信息。
一种示例实施方式中，所述待匹配图像对应的标定框信息中包括多个标定框分别对应的标定框信息；
所述第一匹配模块，包括：
第二确定子模块，设置为从所述多个标定框分别对应的标定框信息中，确定未被标记预设标识的标定框的标定框信息；其中，所述预设标识用于表征对应的标定框已匹配成功；
第三匹配子模块，设置为基于所述未被标记预设标识的标定框的标定框信息，对所述待匹配框对应的待匹配信息进行匹配，并在匹配成功时，为与所述待匹配框对应的待匹配信息匹配成功的标定框标记所述预设标识；
第三确定子模块，设置为基于所述多个标定框针对所述预设标识的标记情况，确定所述标定框信息的匹配结果。
一种示例实施方式中，所述第三确定子模块，设置为：
如果确定所述多个标定框均标记有所述预设标识，则确定所述标定框信息的匹配结果为匹配成功；如果确定所述多个标定框中的至少一个标定框未标记有所述预设标识，确定所述标定框信息的匹配结果为匹配失败。
一种示例实施方式中，所述待匹配图像属于预设第一时间段内连续帧图像中的一帧图像；所述装置还包括：
第二确定模块，设置为在确定所述待匹配图像的匹配结果为匹配成功时，将所述待匹配图像确定为所述预设第一时间段对应的匹配结果图像；
第一更新模块，设置为在确定所述待匹配图像的匹配结果为匹配失败时，基于在所述预设第一时间段内的所述待匹配图像的下一帧图像更新所述待匹配图像，继续触发所述第一识别模块，直到确定所述待匹配图像的匹配结果为匹配成功。
一种示例实施方式中，所述预设第一时间段属于多个预设时间段中的一个，所述装置还包括：
第三确定模块，设置为基于所述多个预设时间段分别对应的匹配结果图像，确定预设个数的目标输出图像；
显示模块，设置为显示所述预设个数的目标输出图像。
一种示例实施方式中，所述第三确定模块，包括：
第四确定子模块，设置为在确定所述预设第一时间段对应的匹配结果图像后，确定所述预设个数的槽位中是否存在空闲槽位；其中，所述预设个数的槽位用于存储所述预设个数的目标输出图像；
第一存入子模块，设置为在确定所述预设个数的槽位中存在空闲槽位时，将所述预设第一时间段对应的匹配结果图像存入所述空闲槽位中；
第一更新子模块，设置为在确定所述预设个数的槽位中不存在空闲槽位时，基于所述匹配结果图像的匹配结果更新所述预设个数的槽位。
一种示例实施方式中，所述第一更新子模块，设置为在确定所述预设个数的槽位中不存在空闲槽位时，从所述匹配结果图像和所述预设个数的槽位中的图像中随机确定所述预设个数的图像，并将所述预设个数的图像分别存储于所述预设个数的槽位中。
本公开实施例提供的图像匹配装置中，首先，确定待匹配图像，并对待匹配图像进行目标对象的识别，如果确定待匹配图像上存在该目标对象，则在待匹配图像上确定包含有该目标对象的待匹配框，并获取该待匹配框对应的待匹配信息。然后，将待匹配框对应的待匹配信息与待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果，然后，基于该匹配结果确定待匹配图像的匹配结果。本公开实施例通过与预先标定的待匹配图像对应的标定框信息进行匹配，确定待匹配图像的匹配结果，基于本公开实施例提供的图像匹配装置，能够丰富图像视频类应用程序产品的功能，提升用户的使用体验。
除了上述方法和装置以外，本公开实施例还提供了一种计算机可读存储介质，计算机可读存储介质中存储有指令，当所述指令在终端设备上运行时，使得所述终端设备实现本公开实施例所述的图像匹配方法。计算机可读存储介质可以是非暂态计算机可读存储介质。
本公开实施例还提供了一种计算机程序产品，包括计算机程序/指令，所述计算机程序/指令被处理器执行时实现本公开实施例所述的图像匹配方法。
另外，本公开实施例还提供了一种图像匹配设备，参见图4所示，可以包括：
处理器401、存储器402、输入装置403和输出装置404。图像匹配设备中的处理器401的数量可以一个或多个，图4中以一个处理器为例。在本公开的一些实施例中，处理器401、存储器402、输入装置403和输出装置404可通过总线或其它方式连接，其中，图4中以通过总线连接为例。
存储器402可设置为存储软件程序以及模块，处理器401通过运行存储在存储器402的软件程序以及模块，从而执行图像匹配设备的多种功能应用以及数据处理。存储器402可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序等。此外，存储器402可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。输入装置403可用于接收输入的数字或字符信息，以及产生与图像匹配设备的用户设置以及功能控制有关的信号输入。
在本实施例中，处理器401会按照如下的指令，将一个或一个以上的应用程序的进程对应的可执行文件加载到存储器402中，并由处理器401来运行存储在存储器402中的应用程序，从而实现上述图像匹配设备的多种功能。
需要说明的是，在本文中，诸如“第一”和“第二”等之类的关系术语仅仅用来将一个 实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。

Claims (14)

  1. 一种图像匹配方法，包括：
    确定待匹配图像，并对所述待匹配图像进行目标对象的识别；
    响应于识别到所述待匹配图像上存在所述目标对象，在所述待匹配图像上确定包含有所述目标对象的待匹配框，并获取所述待匹配框对应的待匹配信息；
    将所述待匹配框对应的待匹配信息与所述待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果；其中，所述标定框信息基于所述待匹配图像对应的标准图像上展示的参考线标定得到；
    基于所述标定框信息的匹配结果，确定所述待匹配图像的匹配结果。
  2. 根据权利要求1所述的方法，其中，所述标定框信息包括基于所述标准图像上展示的所述参考线标定得到的标定框的标定位置信息，所述待匹配框对应的待匹配信息包括所述待匹配框的待匹配位置信息；
    所述将所述待匹配框对应的待匹配信息与所述待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果，包括：
    获取所述待匹配图像对应的标定框信息中的所述标定位置信息；
    将所述待匹配框的待匹配位置信息与所述标定位置信息进行匹配，得到所述标定位置信息对应的标定框的位置匹配结果；
    基于所述标定框的位置匹配结果，确定所述待匹配图像对应的标定框信息的匹配结果。
  3. 根据权利要求2所述的方法，其中，所述标定框信息还包括预先为所述标定框配置的标定关键点信息，所述待匹配框对应的待匹配信息包括所述待匹配框中的待匹配关键点信息；
    所述将所述待匹配框对应的待匹配信息与所述标定位置信息进行匹配，得到所述标定位置信息对应的标定框的位置匹配结果之后，还包括：
    响应于确定所述位置匹配结果为匹配成功，将所述待匹配框中的待匹配关键点信息与所述标定框对应的标定关键点信息进行匹配，得到所述标定框的关键点匹配结果；
    所述基于所述标定框的位置匹配结果，确定所述待匹配图像对应的标定框信息的匹配结果，包括：
    基于所述标定框的关键点匹配结果，确定所述待匹配图像对应的标定框信息的匹配结果。
  4. 根据权利要求3所述的方法，其中，所述目标对象包括手部和脚部中的至少之一，所述标定关键点信息包括手部关键点信息和脚部关键点信息中的至少之一。
  5. 根据权利要求1所述的方法，其中，所述待匹配图像对应的标定框信息中包括多个标定框分别对应的标定框信息；
    所述将所述待匹配框对应的待匹配信息与所述待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果，包括：
    从所述多个标定框分别对应的标定框信息中，确定未被标记预设标识的标定框的标定框信息；其中，所述预设标识用于表征对应的标定框已匹配成功；
    基于所述未被标记预设标识的标定框的标定框信息，对所述待匹配框对应的待匹配信息进行匹配，响应于确定匹配成功，为与所述待匹配框对应的待匹配信息匹配成功的标定框标记所述预设标识；
    基于所述多个标定框针对所述预设标识的标记情况，确定所述标定框信息的匹配结果。
  6. 根据权利要求5所述的方法，其中，所述基于所述多个标定框针对所述预设标识的标 记情况，确定所述标定框信息的匹配结果，包括：
    响应于确定所述多个标定框分别标记有所述预设标识，确定所述标定框信息的匹配结果为匹配成功；响应于确定所述多个标定框中的至少一个标定框未标记有所述预设标识，确定所述标定框信息的匹配结果为匹配失败。
  7. 根据权利要求1所述的方法，所述待匹配图像属于预设第一时间段内连续帧图像中的一帧图像；所述基于所述标定框信息的匹配结果，确定所述待匹配图像的匹配结果之后，还包括：
    响应于确定所述待匹配图像的匹配结果为匹配成功，将所述待匹配图像确定为所述预设第一时间段对应的匹配结果图像；
    响应于确定所述待匹配图像的匹配结果为匹配失败，基于在所述预设第一时间段内的所述待匹配图像的下一帧图像更新所述待匹配图像，继续触发执行所述对所述待匹配图像进行目标对象的识别步骤，直到确定所述待匹配图像的匹配结果为匹配成功。
  8. 根据权利要求7所述的方法，其中，所述预设第一时间段属于多个预设时间段中的一个，所述方法还包括：
    基于所述多个预设时间段分别对应的匹配结果图像，确定预设个数的目标输出图像；以及，
    显示所述预设个数的目标输出图像。
  9. 根据权利要求8所述的方法，其中，所述基于所述多个预设时间段分别对应的匹配结果图像，确定预设个数的目标输出图像，包括：
    在确定所述预设第一时间段对应的匹配结果图像后，确定所述预设个数的槽位中是否存在空闲槽位；其中，所述预设个数的槽位用于存储所述预设个数的目标输出图像；
    响应于确定所述预设个数的槽位中存在空闲槽位，将所述预设第一时间段对应的匹配结果图像存入所述空闲槽位中；
    响应于确定所述预设个数的槽位中不存在空闲槽位，基于所述匹配结果图像的匹配结果更新所述预设个数的槽位。
  10. 根据权利要求9所述的方法，其中，所述基于所述匹配结果图像的匹配结果更新所述预设个数的槽位，包括：
    从所述匹配结果图像和所述预设个数的槽位中的图像中随机确定所述预设个数的图像，并将所述预设个数的图像分别存储于所述预设个数的槽位中。
  11. 一种图像匹配装置，包括：
    第一识别模块，设置为确定待匹配图像，并对所述待匹配图像进行目标对象的识别；
    第一获取模块，设置为响应于识别到所述待匹配图像上存在所述目标对象，在所述待匹配图像上确定包含有所述目标对象的待匹配框，并获取所述待匹配框对应的待匹配信息；
    第一匹配模块，设置为将所述待匹配框对应的待匹配信息与所述待匹配图像对应的标定框信息进行匹配，得到所述标定框信息的匹配结果；其中，所述标定框信息基于所述待匹配图像对应的标准图像上展示的参考线标定得到；
    第一确定模块，设置为基于所述标定框信息的匹配结果，确定所述待匹配图像的匹配结果。
  12. 一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当所述指令在 终端设备上运行时，使得所述终端设备实现如权利要求1-10任一项所述的方法。
  13. 一种设备，包括：存储器，处理器，及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时，实现如权利要求1-10任一项所述的方法。
  14. 一种计算机程序产品，所述计算机程序产品包括计算机程序/指令，所述计算机程序/指令被处理器执行时使得所述处理器实现如权利要求1-10任一项所述的方法。
PCT/CN2022/105448 2021-07-19 2022-07-13 图像匹配方法、装置、设备及存储介质 Ceased WO2023001039A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/569,666 US20240282142A1 (en) 2021-07-19 2022-07-13 Image matching method and apparatus, and device and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110812807.5A CN115701873A (zh) 2021-07-19 2021-07-19 一种图像匹配方法、装置、设备及存储介质
CN202110812807.5 2021-07-19

Publications (1)

Publication Number Publication Date
WO2023001039A1 true WO2023001039A1 (zh) 2023-01-26

Family

ID=84978868

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/105448 Ceased WO2023001039A1 (zh) 2021-07-19 2022-07-13 图像匹配方法、装置、设备及存储介质

Country Status (3)

Country Link
US (1) US20240282142A1 (zh)
CN (1) CN115701873A (zh)
WO (1) WO2023001039A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108682021A (zh) * 2018-04-18 2018-10-19 平安科技（深圳）有限公司 快速手部跟踪方法、装置、终端及存储介质
CN111985385A (zh) * 2020-08-14 2020-11-24 杭州海康威视数字技术股份有限公司 一种行为检测方法、装置及设备
CN111986252A (zh) * 2020-07-16 2020-11-24 浙江工业大学 一种用于目标分割网络中候选边界框精准定位的方法
CN112184802A (zh) * 2019-07-05 2021-01-05 杭州海康威视数字技术股份有限公司 标定框的调整方法、装置及存储介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108874120A (zh) * 2018-03-29 2018-11-23 北京字节跳动网络技术有限公司 人机交互系统、方法、计算机可读存储介质及交互装置
CN109194879B (zh) * 2018-11-19 2021-09-07 Oppo广东移动通信有限公司 拍照方法、装置、存储介质及移动终端
CN110007765A (zh) * 2019-04-11 2019-07-12 上海星视度科技有限公司 一种人机交互方法、装置及设备
CN110177206A (zh) * 2019-05-27 2019-08-27 努比亚技术有限公司 拍摄方法、移动终端及计算机可读存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108682021A (zh) * 2018-04-18 2018-10-19 平安科技（深圳）有限公司 快速手部跟踪方法、装置、终端及存储介质
CN112184802A (zh) * 2019-07-05 2021-01-05 杭州海康威视数字技术股份有限公司 标定框的调整方法、装置及存储介质
CN111986252A (zh) * 2020-07-16 2020-11-24 浙江工业大学 一种用于目标分割网络中候选边界框精准定位的方法
CN111985385A (zh) * 2020-08-14 2020-11-24 杭州海康威视数字技术股份有限公司 一种行为检测方法、装置及设备

Also Published As

Publication number Publication date
CN115701873A (zh) 2023-02-14
US20240282142A1 (en) 2024-08-22

Similar Documents

Publication Publication Date Title
BR112016017262B1 (pt) Método para busca de objeto e terminal acoplado de forma comunicativa a um servidor.
JP7390628B2 (ja) 識別情報付与装置、識別情報付与方法、及びプログラム
CN111324352A (zh) 一种应用页面的代码生成方法及相关设备
CN106202071A (zh) 账户信息获取的方法、终端、服务器和系统
CN112101353B (zh) 文本信息提取方法、装置、电子设备及存储介质
JP2018033365A (ja) 情報処理装置、情報処理方法、プログラム
CN110431838A (zh) 提供人脸识别摄像机的动态内容的方法及系统
JP2024525281A5 (zh)
US10438085B2 (en) Image analysis apparatus, image analysis method, and storage medium
CN107122093B (zh) 信息框显示方法及装置
CN109871205B (zh) 界面代码调整方法、装置、计算机装置及存储介质
WO2019119643A1 (zh) 移动直播的互动终端、方法及计算机可读存储介质
CN112287850B (zh) 物品信息识别方法、装置、电子设备和可读存储介质
WO2023001039A1 (zh) 图像匹配方法、装置、设备及存储介质
CN114782994A (zh) 姿态识别方法、装置、存储介质及电子设备
CN118860127A (zh) 数据处理方法、电子设备、计算机存储介质及程序产品
CN108419118A (zh) 电视机用户界面的生成方法、电视机及可读存储介质
US12608965B2 (en) Electronic device for generating clothing management information and method of controlling the same
US20240406339A1 (en) Video generation method and apparatus, and device and storage medium
CN112954666B (zh) 一种用户画像方法、装置、设备和存储介质
CN110569082B (zh) 对页面内容进行控制的方法及装置
CN117274505B (zh) 图像映射到搭建图的显示方法及装置
WO2016173136A1 (zh) 一种终端应用处理方法及其装置
CN115457649B (zh) 步态识别方法、电子设备和计算机可读存储介质
CN111652182B (zh) 一种悬空手势识别的方法、装置、电子设备和存储介质

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 18569666

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

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

Ref document number: 22845204

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 22845204

Country of ref document: EP

Kind code of ref document: A1