Motion capture prediction optimization method and system in three-dimensional manufacturing

Abstract

The invention is suitable for the technical field of three-dimensional manufacturing, and provides a motion capture prediction optimization method and system in three-dimensional manufacturing. According to the invention, by tracking and monitoring the positioning identifier of the action target, the basic shooting route is optimized in real time; performing tracking binocular shooting on the motion target; performing position identification on the plurality of active capturing identifiers and the corresponding auxiliary capturing identifiers of the action target; performing offset identification on the identification space data; and according to the relative offset data and the back-to-back offset data, calculating relative optimization deviation and back-to-back optimization deviation, and carrying out motion capture prediction optimization processing on the plurality of active capture identifiers. A shooting route can be optimized to perform tracking binocular shooting, position identification is performed on a plurality of active capture identifiers and corresponding auxiliary capture identifiers, then offset identification and deviation optimization are performed according to opposite and back-to-back aspects, and motion capture prediction optimization processing is performed, so that identifier offset optimization is realized, motion capture accuracy is effectively improved, and motion capture efficiency is improved. Therefore, a vivid action animation effect can be achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of 3D production, and particularly relates to a method and system for optimizing motion capture prediction in 3D production. Background Art

[0002] 3D production is a technology for creating, editing, and optimizing 3D data models in a virtual 3D space by using professional software and technologies. These models can be static objects or dynamic scenes or characters. For the 3D production of dynamic characters, motion capture is sometimes required to produce the desired motion animation effects.

[0003] In the prior art, motion capture in 3D production mainly uses optical motion capture technology, which is completed by tracking and monitoring the markers pasted on the motion target. However, due to the influence of factors such as possible clothing wrinkles and air disturbances on the outer surface of the motion target, it is easy for the markers to shift, affecting the accuracy of motion capture and making the motion animation effects in 3D production inconsistent with the actual situation. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a method and system for optimizing motion capture prediction in 3D production, aiming to solve the technical problems existing in the prior art mentioned in the background art.

[0005] The embodiments of the present invention are implemented as follows: A method for optimizing motion capture prediction in 3D production, the method specifically includes the following steps: Obtain the basic movement route of the motion target, plan the basic shooting route, and track and monitor the positioning markers of the motion target, and optimize the basic shooting route in real time to generate an optimized shooting route; According to the optimized shooting route, perform binocular shooting on the motion target to obtain binocular shooting data in real time; According to the binocular shooting data, identify the positions of multiple active motion capture markers and corresponding auxiliary motion capture markers of the motion target to obtain marker space data; Perform offset identification on the marker space data, select multiple relatively offset markers and multiple oppositely offset markers, and extract relative offset data and oppositely offset data; According to the relative offset data and the oppositely offset data, calculate the relative optimization deviation and the oppositely optimization deviation, and perform motion capture prediction optimization processing on multiple active motion capture markers.

[0006] As a further limitation of the technical solution of the embodiment of the present invention, the steps of obtaining the basic movement route of the action target, planning the basic shooting route, tracking and monitoring the positioning identifier of the action target, and optimizing the basic shooting route in real time to generate the optimized shooting route specifically include the following steps: Obtain the basic movement route of the action target and the motion capture shooting parameters; According to the basic movement route and the motion capture shooting parameters, perform motion capture shooting planning to generate a basic shooting route; Determine the positioning identifier of the action target; During the movement of the action target, track and monitor the positioning identifier of the action target to obtain positioning monitoring data; According to the positioning monitoring data, optimize the basic shooting route in real time to generate the optimized shooting route.

[0007] As a further limitation of the technical solution of the embodiment of the present invention, the steps of performing tracking binocular shooting on the action target according to the optimized shooting route and obtaining binocular shooting data in real time specifically include the following steps: Perform shooting tracking control according to the optimized shooting route; Determine the motion capture shooting frame rate; According to the motion capture shooting frame rate, perform binocular shooting on the action target to obtain binocular shooting data in real time.

[0008] As a further limitation of the technical solution of the embodiment of the present invention, the steps of performing position recognition on multiple active motion capture identifiers and corresponding auxiliary motion capture identifiers of the action target according to the binocular shooting data to obtain identifier space data specifically include the following steps: Obtain the identifier feature data of multiple active motion capture identifiers and corresponding auxiliary motion capture identifiers; Based on the identifier feature data, perform feature matching on the binocular shooting data and record the feature matching information; According to the feature matching information, perform position recognition on multiple active motion capture identifiers and corresponding auxiliary motion capture identifiers to obtain identifier space data.

[0009] As a further limitation of the technical solution of the embodiment of the present invention, the steps of performing offset recognition on the identifier space data, selecting multiple relative offset identifiers and multiple opposite offset identifiers, and extracting relative offset data and opposite offset data specifically include the following steps: According to the identifier space data, record the relative identifier distances between multiple active motion capture identifiers and corresponding auxiliary motion capture identifiers; Based on a preset standard distance and tolerance interval, perform offset comparison on multiple relative identifier distances and record the offset comparison results; According to the offset comparison result, select a plurality of relative offset identifiers and a plurality of opposite offset identifiers from the plurality of active capture identifiers; Extract relative offset data and opposite offset data from the plurality of relative identifier distances.

[0010] As a further limitation of the technical solution of the embodiment of the present invention, the calculating relative optimization deviation and opposite optimization deviation according to the relative offset data and the opposite offset data, and performing dynamic capture prediction optimization processing on the plurality of active capture identifiers specifically includes the following steps: Calculate relative optimization deviation and opposite optimization deviation according to the relative offset data and the opposite offset data; Perform spatial deviation analysis on the identifier space data to determine the relative deviation direction of the plurality of relative offset identifiers and the opposite deviation direction of the plurality of opposite offset identifiers; Perform dynamic capture prediction optimization processing on the plurality of corresponding relative offset identifiers according to the relative optimization deviation and the plurality of relative deviation directions; Perform dynamic capture prediction optimization processing on the plurality of corresponding opposite offset identifiers according to the opposite optimization deviation and the plurality of opposite deviation directions.

[0011] As a further limitation of the technical solution of the embodiment of the present invention, the calculation formula of the relative optimization deviation is: ; wherein, is the relative optimization deviation, is the relative identifier distance between the th relative offset identifier and the corresponding auxiliary dynamic capture identifier, and there are relative offset identifiers in total, is the standard distance; The calculation formula of the opposite optimization deviation is: ; wherein, is the opposite optimization deviation, is the relative identifier distance between the th opposite offset identifier and the corresponding auxiliary dynamic capture identifier, and there are opposite offset identifiers in total.

[0012] A dynamic capture prediction optimization system in 3D production, the system includes a shooting route optimization module, a tracking binocular shooting module, an identifier position recognition module, an identifier offset recognition module, and a dynamic capture prediction optimization module, wherein: The shooting route optimization module is used to obtain the basic movement route of the action target, plan the basic shooting route, track and monitor the positioning identifier of the action target, and optimize the basic shooting route in real time to generate an optimized shooting route; The tracking binocular shooting module is used to perform tracking binocular shooting on the action target according to the optimized shooting route and obtain binocular shooting data in real time; The identification position recognition module is used to perform position recognition on multiple active capture identifiers and corresponding auxiliary capture identifiers of the action target according to the binocular shooting data to obtain identification space data; The identification offset recognition module is used to perform offset recognition on the identification space data, select multiple relative offset identifiers and multiple opposite offset identifiers, and extract relative offset data and opposite offset data; The motion capture prediction optimization module is used to calculate the relative optimization deviation and the opposite optimization deviation according to the relative offset data and the opposite offset data, and perform motion capture prediction optimization processing on multiple active capture identifiers.

[0013] As a further limitation of the technical solution of the embodiment of the present invention, the shooting route optimization module specifically includes: The basic acquisition unit is used to obtain the basic movement route of the action target and the motion capture shooting parameters; The shooting planning unit is used to perform motion capture shooting planning according to the basic movement route and the motion capture shooting parameters to generate a basic shooting route; The identifier determination unit is used to determine the positioning identifier of the action target; The tracking and monitoring unit is used to track and monitor the positioning identifier of the action target during the movement of the action target to obtain positioning monitoring data; The route optimization unit is used to optimize the basic shooting route in real time according to the positioning monitoring data to generate an optimized shooting route.

[0014] As a further limitation of the technical solution of the embodiment of the present invention, the identification offset recognition module specifically includes: The distance recording unit is used to record the relative identifier distances between multiple active capture identifiers and corresponding auxiliary capture identifiers according to the identification space data; The offset comparison unit is used to perform offset comparison on multiple relative identifier distances based on a preset standard distance and a tolerance interval, and record the offset comparison results; The identifier selection unit is used to select multiple relative offset identifiers and multiple opposite offset identifiers from multiple active capture identifiers according to the offset comparison results; The data extraction unit is used to extract relative offset data and opposite offset data from multiple relative identifier distances.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: In the embodiment of the present invention, by tracking and monitoring the positioning identification of the action target, the basic shooting route is optimized in real time; binocular shooting is performed on the action target; the positions of multiple active capture identifiers and corresponding auxiliary capture identifiers of the action target are identified; offset identification is performed on the identifier spatial data; according to the relative offset data and the opposite offset data, the relative optimization deviation and the opposite optimization deviation are calculated, and motion capture prediction optimization processing is performed on multiple active capture identifiers. It is possible to optimize the shooting route for binocular tracking shooting, identify the positions of multiple active capture identifiers and corresponding auxiliary capture identifiers, then perform offset identification and deviation optimization in terms of relative and opposite aspects, and then perform motion capture prediction optimization processing, thereby realizing the optimization of identifier offset, effectively improving the accuracy of motion capture, and creating a realistic motion animation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Shows a flowchart of a motion capture prediction optimization method in 3D production provided by an embodiment of the present invention; Figure 2 Shows a flowchart of real-time optimization of the basic shooting route in the method provided by an embodiment of the present invention; Figure 3 Shows a flowchart of performing binocular tracking shooting in the method provided by an embodiment of the present invention; Figure 4 Shows a flowchart of obtaining identifier spatial data in the method provided by an embodiment of the present invention; Figure 5 Shows a flowchart of offset identification and data extraction in the method provided by an embodiment of the present invention; Figure 6 Shows a flowchart of performing motion capture prediction optimization processing in the method provided by an embodiment of the present invention; Figure 7 Shows an application architecture diagram of a motion capture prediction optimization system in 3D production provided by an embodiment of the present invention; Figure 8 Shows a structural block diagram of a shooting route optimization module in the system provided by an embodiment of the present invention; Figure 9 Shows a structural block diagram of an identifier offset identification module in the system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0018] It can be understood that in the prior art, for motion capture in 3D production, mainly optical motion capture technology is adopted, which is completed by tracking and monitoring the markers pasted on the motion target. However, due to the influence of factors such as possible clothing wrinkles and air disturbances on the outer surface of the motion target, it is easy for the markers to shift, affecting the accuracy of motion capture, and making the 3D-produced motion animation effect inconsistent with the actual situation.

[0019] To solve the above problems, a motion capture prediction optimization method and system in 3D production disclosed in an embodiment of the present invention obtain the basic movement route of a motion target, plan a basic shooting route, track and monitor the positioning markers of the motion target, and optimize the basic shooting route in real time to generate an optimized shooting route; according to the optimized shooting route, perform tracking binocular shooting on the motion target to obtain binocular shooting data in real time; based on the binocular shooting data, perform position recognition on multiple active motion capture markers and corresponding auxiliary motion capture markers of the motion target to obtain marker space data; perform offset recognition on the marker space data, select multiple relatively offset markers and multiple oppositely offset markers, and extract relatively offset data and oppositely offset data; according to the relatively offset data and the oppositely offset data, calculate relatively optimized deviation and oppositely optimized deviation, and perform motion capture prediction optimization processing on multiple active motion capture markers. It can optimize the shooting route for tracking binocular shooting, perform position recognition on multiple active motion capture markers and corresponding auxiliary motion capture markers, then perform offset recognition and deviation optimization from two aspects of relative and opposite, and further perform motion capture prediction optimization processing, so as to realize the optimization of marker offset, effectively improve the accuracy of motion capture, and produce a realistic motion animation effect.

[0020] Specifically, Figure 1 The flowchart of the motion capture prediction optimization method in 3D production provided by an embodiment of the present invention is shown.

[0021] In a preferred embodiment provided by the present invention, a motion capture prediction optimization method in 3D production, the method specifically includes the following steps: Step S101, obtain the basic movement route of the motion target, plan the basic shooting route, track and monitor the positioning markers of the motion target, and optimize the basic shooting route in real time to generate an optimized shooting route.

[0022] In an embodiment of the present invention, by obtaining the basic movement route of the action target and the motion capture shooting parameters, and then based on the basic movement route and the motion capture shooting parameters, performing motion capture shooting planning to ensure that during the shooting process, it can be synchronized with the basic movement route of the action target, and always maintain a separation distance set in the motion capture shooting parameters from the action target, and generating a basic shooting route according to the shooting height and shooting angle. By determining the positioning identifier of the action target, during the movement of the action target, continuously performing infrared monitoring on the action target and conducting positioning analysis to obtain the positioning monitoring data of the action target. By comparing the positioning monitoring data with the basic movement route for coincidence, it is determined whether the movement process of the action target deviates from the basic movement route. When the action target deviates from the basic movement route, based on the current position deviation, keeping the subsequent movement route of the action target unchanged, and making corresponding adjustments to the basic shooting route to generate an optimized shooting route.

[0023] It can be understood that the motion capture shooting parameters include parameters such as shooting distance, shooting height, shooting angle, and shooting frame rate.

[0024] It can be understood that the positioning identifier can be an infrared identifier that maintains a fixed temperature and has an obvious difference from the temperature of the action target. During the movement of the action target, infrared monitoring can be performed on the positioning identifier to achieve the positioning and identification of the action target.

[0025] Specifically, Figure 2 The flowchart showing the real-time optimization of the basic shooting route in the method provided by the embodiment of the present invention is presented.

[0026] Among them, in another preferred embodiment provided by the present invention, the steps of obtaining the basic movement route of the action target, planning the basic shooting route, tracking and monitoring the positioning identifier of the action target, and real-time optimizing the basic shooting route to generate an optimized shooting route specifically include the following steps: Step S1011: Obtain the basic movement route of the action target and the motion capture shooting parameters.

[0027] Step S1012: Based on the basic movement route and the motion capture shooting parameters, perform motion capture shooting planning to generate a basic shooting route.

[0028] Step S1013: Determine the positioning identifier of the action target.

[0029] Step S1014: During the movement of the action target, track and monitor the positioning identifier of the action target to obtain positioning monitoring data.

[0030] Step S1015: Based on the positioning monitoring data, real-time optimize the basic shooting route to generate an optimized shooting route.

[0031] Further, the motion capture prediction optimization method in the 3D production further includes the following steps: Step S102: Track and binocularly shoot the action target according to the optimized shooting route, and obtain binocular shooting data in real time.

[0032] In the embodiment of the present invention, during the movement of the action target, according to the optimized shooting route, shooting tracking control is performed on the action target, so that during the shooting process, it can move according to the optimized shooting route, and real-time adjustment is performed during the real-time optimization of the route, and the motion capture shooting frame rate is determined. According to the motion capture shooting frame rate, binocular shooting and transmission are performed on the action target, and binocular shooting data is obtained in real time.

[0033] Specifically, Figure 3 The flowchart of binocular shooting for tracking in the method provided by the embodiment of the present invention is shown.

[0034] Wherein, in another preferred embodiment provided by the present invention, the step of tracking and binocularly shooting the action target according to the optimized shooting route and obtaining binocular shooting data in real time specifically includes the following steps: Step S1021: Perform shooting tracking control according to the optimized shooting route.

[0035] Step S1022: Determine the motion capture shooting frame rate.

[0036] Step S1023: Binocularly shoot the action target according to the motion capture shooting frame rate, and obtain binocular shooting data in real time.

[0037] Further, the motion capture prediction optimization method in the 3D production further includes the following steps: Step S103: Identify the positions of multiple active motion capture identifiers and corresponding auxiliary motion capture identifiers of the action target according to the binocular shooting data, and obtain identifier space data.

[0038] In the embodiment of the present invention, by obtaining the identifier feature data of multiple active motion capture identifiers and corresponding auxiliary motion capture identifiers, then based on the identifier feature data, feature matching of the identifiers is performed on the binocular shooting data, and the matching results are recorded to generate feature matching information. Then, according to the feature matching information, the spatial positions of multiple active motion capture identifiers and corresponding auxiliary motion capture identifiers are identified and recorded to obtain identifier space data.

[0039] It can be understood that multiple active motion capture identifiers are pasted on the outer surface of the action target, and corresponding auxiliary motion capture identifiers are pasted at a standard distance from each active motion capture identifier, and the active motion capture identifier and the corresponding auxiliary motion capture identifier need to be on the same motion joint.

[0040] It can be understood that in the embodiments of the present invention, a point can be selected as the coordinate origin in the positioning identifier to construct a spatial coordinate system, and then in the spatial coordinate system, the spatial positions of multiple active capture identifiers and corresponding passive capture identifiers can be identified.

[0041] Specifically, Figure 4 FIG. shows a flowchart of obtaining identifier spatial data in the method provided by the embodiments of the present invention.

[0042] Among them, in another preferred embodiment provided by the present invention, the position recognition of multiple active capture identifiers and corresponding passive capture identifiers of the action target according to the binocular shooting data to obtain identifier spatial data specifically includes the following steps: Step S1031: Obtain the identifier feature data of multiple active capture identifiers and corresponding passive capture identifiers.

[0043] Step S1032: Based on the identifier feature data, perform feature matching on the binocular shooting data and record the feature matching information.

[0044] Step S1033: According to the feature matching information, perform position recognition on multiple active capture identifiers and corresponding passive capture identifiers to obtain identifier spatial data.

[0045] Further, the dynamic capture prediction optimization method in the 3D production further includes the following steps: Step S104: Perform offset recognition on the identifier spatial data, select multiple relative offset identifiers and multiple opposite offset identifiers, and extract relative offset data and opposite offset data.

[0046] In the embodiments of the present invention, according to the spatial coordinate data of multiple active capture identifiers and corresponding passive capture identifiers in the identifier spatial data, the relative identifier distance between multiple active capture identifiers and corresponding passive capture identifiers is calculated, and then with a preset standard distance and error tolerance interval, a comparative analysis of the offset of multiple relative identifier distances is performed. When the relative identifier distance is less than the standard distance and the difference between the relative identifier distance and the standard distance is greater than the error tolerance interval, the corresponding active capture identifier is a relative offset identifier; when the relative identifier distance is greater than the standard distance and the difference between the relative identifier distance and the standard distance is greater than the error tolerance interval, the corresponding active capture identifier is an opposite offset identifier, so that multiple relative offset identifiers and multiple opposite offset identifiers can be selected from multiple active capture identifiers, and then from multiple relative identifier distances, the relative offset data of multiple relative offset identifiers is extracted, and the opposite offset data of multiple opposite offset identifiers is extracted.

[0047] Specifically, Figure 5 FIG. shows a flowchart of offset recognition and data extraction in the method provided by the embodiments of the present invention.

[0048] Among them, in another preferred embodiment provided by the present invention, the offset recognition of the identification space data, the selection of a plurality of relative offset identifiers and a plurality of opposite offset identifiers, and the extraction of relative offset data and opposite offset data specifically include the following steps: Step S1041: According to the identification space data, record the relative identification distances between a plurality of active capture identifiers and the corresponding auxiliary capture identifiers.

[0049] Step S1042: Based on a preset standard distance and a tolerance interval, perform offset comparison on the plurality of relative identification distances, and record the offset comparison results.

[0050] Step S1043: According to the offset comparison results, select a plurality of relative offset identifiers and a plurality of opposite offset identifiers from the plurality of active capture identifiers.

[0051] Step S1044: Extract relative offset data and opposite offset data from the plurality of relative identification distances.

[0052] Furthermore, the motion capture prediction optimization method in the 3D production further includes the following steps: Step S105: According to the relative offset data and the opposite offset data, calculate the relative optimization deviation and the opposite optimization deviation, and perform motion capture prediction optimization processing on the plurality of active capture identifiers.

[0053] In the embodiments of the present invention, according to the relative offset data and the opposite offset data, calculate the relative optimization deviation and the opposite optimization deviation, and then perform spatial deviation analysis on the identification space data to determine, in the spatial coordinate system, the relative deviation directions of a plurality of relative offset identifiers approaching the corresponding auxiliary capture identifiers, and the opposite deviation directions of a plurality of opposite offset identifiers moving away from the corresponding auxiliary capture identifiers. Furthermore, perform motion capture prediction optimization processing on the plurality of corresponding relative offset identifiers in the direction opposite to the relative optimization deviation and the plurality of relative deviation directions, and perform motion capture prediction optimization processing on the plurality of corresponding opposite offset identifiers in the direction opposite to the opposite optimization deviation and the plurality of opposite deviation directions. Specifically, the calculation formula for the relative optimization deviation is: ; Wherein, is the relative optimization deviation, is the th relative identification distance between the relative offset identifier and the corresponding auxiliary capture identifier, and there are relative offset identifiers in total, is the standard distance; The calculation formula for the opposite optimization deviation is: ; Wherein, is the opposite optimization deviation, For the relative identification distance between the nth opposite offset identification and the corresponding auxiliary motion capture identification, there are opposite offset identifications.

[0054] Specifically, Figure 6 The flowchart of performing motion capture prediction optimization processing in the method provided by the embodiment of the present invention is shown.

[0055] Wherein, in another preferred embodiment provided by the present invention, calculating the relative optimization deviation and the opposite optimization deviation according to the relative offset data and the opposite offset data, and performing motion capture prediction optimization processing on a plurality of the active capture identifications specifically includes the following steps: Step S1051: Calculate the relative optimization deviation and the opposite optimization deviation according to the relative offset data and the opposite offset data.

[0056] Step S1052: Perform spatial deviation analysis on the identification space data to determine the relative deviation directions of a plurality of the relative offset identifications and the opposite deviation directions of a plurality of the opposite offset identifications.

[0057] Step S1053: Perform motion capture prediction optimization processing on a plurality of corresponding relative offset identifications according to the relative optimization deviation and a plurality of the relative deviation directions.

[0058] Step S1054: Perform motion capture prediction optimization processing on a plurality of corresponding opposite offset identifications according to the opposite optimization deviation and a plurality of the opposite deviation directions.

[0059] Furthermore, Figure 7 The application architecture diagram of the motion capture prediction optimization system in 3D production provided by the embodiment of the present invention is shown.

[0060] Specifically, in another preferred embodiment provided by the present invention, a motion capture prediction optimization system in 3D production includes: A shooting route optimization module 101, configured to obtain the basic movement route of an action target, plan the basic shooting route, and perform tracking and monitoring on the positioning identification of the action target, and optimize the basic shooting route in real time to generate an optimized shooting route.

[0061] In an embodiment of the present invention, the shooting route optimization module 101 obtains the basic movement route of the action target and the motion capture shooting parameters, and then performs motion capture shooting planning according to the basic movement route and the motion capture shooting parameters to ensure that during the shooting process, it can keep in sync with the basic movement route of the action target, and always maintain a separation distance set in the motion capture shooting parameters from the action target, and generate a basic shooting route according to the shooting height and shooting angle. By determining the positioning identifier of the action target, during the movement process of the action target, continuous infrared monitoring is performed on the action target, and positioning analysis is carried out to obtain the positioning monitoring data of the action target. By comparing the positioning monitoring data with the basic movement route for coincidence, it is judged whether the movement process of the action target deviates from the basic movement route. When the action target deviates from the basic movement route, based on the current position deviation, the subsequent movement route of the action target remains unchanged, and the basic shooting route is adjusted accordingly to generate an optimized shooting route.

[0062] Specifically, Figure 8 FIG. shows the structural block diagram of the shooting route optimization module 101 in the system provided by the embodiment of the present invention.

[0063] Among them, in another preferred embodiment provided by the present invention, the shooting route optimization module 101 specifically includes: A basic acquisition unit 1011, configured to obtain the basic movement route of the action target and the motion capture shooting parameters.

[0064] A shooting planning unit 1012, configured to perform motion capture shooting planning according to the basic movement route and the motion capture shooting parameters to generate a basic shooting route.

[0065] An identifier determination unit 1013, configured to determine the positioning identifier of the action target.

[0066] A tracking and monitoring unit 1014, configured to perform tracking and monitoring on the positioning identifier of the action target during the movement process of the action target to obtain positioning monitoring data.

[0067] A route optimization unit 1015, configured to optimize the basic shooting route in real time according to the positioning monitoring data to generate an optimized shooting route.

[0068] Furthermore, the motion capture prediction optimization system in the 3D production further includes: A tracking binocular shooting module 102, configured to perform tracking binocular shooting on the action target according to the optimized shooting route to obtain binocular shooting data in real time.

[0069] In an embodiment of the present invention, during the movement of the action target, the tracking binocular shooting module 102 controls the shooting and tracking of the action target according to the optimized shooting route, so that during the shooting process, it can move according to the optimized shooting route, and perform real-time adjustment during the real-time optimization of the route, and determine the motion capture shooting frame rate. According to the motion capture shooting frame rate, binocular shooting and transmission of the action target are performed to obtain binocular shooting data in real time.

[0070] The identification position recognition module 103 is configured to perform position recognition on multiple active capture identifiers and corresponding auxiliary active capture identifiers of the action target according to the binocular shooting data, and obtain identification space data.

[0071] In an embodiment of the present invention, the identification position recognition module 103 obtains the identification feature data of multiple active capture identifiers and corresponding auxiliary active capture identifiers, and then based on the identification feature data, performs feature matching of the identifiers on the binocular shooting data, records the matching results, generates feature matching information, and then according to the feature matching information, performs spatial position recognition and recording of multiple active capture identifiers and corresponding auxiliary active capture identifiers to obtain identification space data.

[0072] The identification offset recognition module 104 is configured to perform offset recognition on the identification space data, select multiple relative offset identifiers and multiple opposite offset identifiers, and extract relative offset data and opposite offset data.

[0073] In an embodiment of the present invention, the identification offset recognition module 104 calculates the relative identification distance between multiple active capture identifiers and corresponding auxiliary active capture identifiers according to the spatial coordinate data of multiple active capture identifiers and corresponding auxiliary active capture identifiers in the identification space data, and then uses a preset standard distance and tolerance interval to perform comparative analysis of the offset of multiple relative identification distances. When the relative identification distance is less than the standard distance and the difference between the relative identification distance and the standard distance is greater than the tolerance interval, the corresponding active capture identifier is a relative offset identifier; when the relative identification distance is greater than the standard distance and the difference between the relative identification distance and the standard distance is greater than the tolerance interval, the corresponding active capture identifier is an opposite offset identifier, so that multiple relative offset identifiers and multiple opposite offset identifiers can be selected from multiple active capture identifiers, and then multiple relative offset data of multiple relative offset identifiers and multiple opposite offset data of multiple opposite offset identifiers can be extracted from multiple relative identification distances.

[0074] Specifically, Figure 9 Fig. shows the structural block diagram of the identification offset recognition module 104 in the system provided by the embodiment of the present invention.

[0075] Wherein, in another preferred embodiment provided by the present invention, the identification offset recognition module 104 specifically includes: A distance recording unit 1041, configured to record a relative identification distance between a plurality of active capture identifiers and corresponding passive capture identifiers according to the identification space data.

[0076] An offset comparison unit 1042, configured to perform an offset comparison on a plurality of the relative identification distances based on a preset standard distance and a tolerance interval, and record an offset comparison result.

[0077] An identification selection unit 1043, configured to select a plurality of relatively offset identifiers and a plurality of oppositely offset identifiers from a plurality of the active capture identifiers according to the offset comparison result.

[0078] A data extraction unit 1044, configured to extract relatively offset data and oppositely offset data from a plurality of the relative identification distances.

[0079] Further, the motion capture prediction optimization system in the 3D production further includes: A motion capture prediction optimization module 105, configured to calculate a relative optimization deviation and an opposite optimization deviation according to the relatively offset data and the oppositely offset data, and perform motion capture prediction optimization processing on a plurality of the active capture identifiers.

[0080] In an embodiment of the present invention, the motion capture prediction optimization module 105 calculates a relative optimization deviation and an opposite optimization deviation according to the relatively offset data and the oppositely offset data, and then performs a spatial deviation analysis on the identification space data to determine, in a spatial coordinate system, a relative deviation direction in which a plurality of relatively offset identifiers approach corresponding passive capture identifiers, and an opposite deviation direction in which a plurality of oppositely offset identifiers move away from corresponding passive capture identifiers. Furthermore, the motion capture prediction optimization processing is performed on a plurality of corresponding relatively offset identifiers in a direction opposite to the relative optimization deviation and a plurality of relative deviation directions, and the motion capture prediction optimization processing is performed on a plurality of corresponding oppositely offset identifiers in a direction opposite to the opposite optimization deviation and a plurality of opposite deviation directions. Specifically, the calculation formula for the relative optimization deviation is: ; Wherein, is the relative optimization deviation, is the relative identification distance between the th relatively offset identifier and the corresponding passive capture identifier, and there are relatively offset identifiers in total, is the standard distance; The calculation formula for the opposite optimization deviation is: ; Wherein, is the opposite optimization deviation, is the relative identification distance between the th oppositely offset identifier and the corresponding passive capture identifier, and there are a back-offset identifier

[0081] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown in sequence according to the indications of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0082] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0083] The above embodiments merely represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.

Claims

1. A motion capture prediction optimization method in three-dimensional production, characterized in that: The method specifically comprises the following steps: Obtaining a basic moving route of the action target, planning a basic shooting route, tracking and monitoring the positioning mark of the action target, optimizing the basic shooting route in real time, and generating an optimized shooting route; According to the optimized shooting route, the action target is tracked and binocularly photographed, and binocular shooting data is acquired in real time; According to the binocular shooting data, position recognition is performed on multiple active motion capture markers and corresponding auxiliary motion capture markers of the motion target to obtain marker space data; Performing offset identification on the identification space data, selecting a plurality of relative offset identifications and a plurality of opposite offset identifications, and extracting relative offset data and opposite offset data; According to the relative offset data and the opposite offset data, the relative optimization deviation and the opposite optimization deviation are calculated, and the motion capture prediction optimization processing is performed on the multiple active capture markers.

2. The motion capture prediction optimization method in three-dimensional production according to claim 1, characterized in that: The steps of obtaining the basic moving route of the action target, planning the basic shooting route, tracking and monitoring the positioning mark of the action target, optimizing the basic shooting route in real time, and generating the optimized shooting route specifically include the following steps: Obtain the basic movement route and motion capture parameters of the action target; Performing motion capture shooting planning according to the basic moving route and the motion capture shooting parameters to generate a basic shooting route; Determine the location identifier of the action target; During the movement of the action target, tracking and monitoring the positioning identifier of the action target to obtain positioning monitoring data; According to the positioning monitoring data, the basic shooting route is optimized in real time to generate an optimized shooting route.

3. The motion capture prediction optimization method in three-dimensional production according to claim 1, characterized in that: The step of tracking the action target by binocular photography according to the optimized photography route and acquiring binocular photography data in real time specifically comprises the following steps: According to the optimized shooting route, shooting tracking control is performed; Determine the frame rate of motion capture; According to the motion capture frame rate, binocular photography is performed on the motion target to obtain binocular photography data in real time.

4. The method for motion capture prediction and optimization in three-dimensional production according to claim 1, characterized in that: The step of performing position recognition on multiple active motion capture markers and corresponding auxiliary motion capture markers of the motion target according to the binocular shooting data to obtain marker space data specifically comprises the following steps: Acquire identification feature data of multiple active capture identifications and corresponding auxiliary capture identifications; Based on the identification feature data, feature matching is performed on the binocular shooting data, and feature matching information is recorded; According to the feature matching information, positions of multiple active motion capture markers and corresponding auxiliary motion capture markers are identified to obtain marker space data.

5. The method for motion capture prediction and optimization in three-dimensional production according to claim 1, characterized in that: The step of performing offset identification on the identification space data, selecting a plurality of relative offset identifications and a plurality of opposite offset identifications, and extracting the relative offset data and the opposite offset data specifically comprises the following steps: According to the identification space data, the relative identification distances between the plurality of active capture identifications and the corresponding auxiliary motion capture identifications are recorded; Based on a preset standard distance and a tolerance interval, performing an offset comparison on a plurality of the relative identification distances, and recording an offset comparison result; According to the offset comparison result, selecting a plurality of relative offset identifiers and a plurality of opposite offset identifiers from the plurality of active capture identifiers; Relative offset data and opposite offset data are extracted from the plurality of relative identification distances.

6. The motion capture prediction optimization method in three-dimensional production according to claim 5, characterized in that: The step of calculating the relative optimization deviation and the opposite optimization deviation according to the relative offset data and the opposite offset data, and performing motion capture prediction optimization processing on the multiple active capture markers specifically includes the following steps: Calculating a relative optimization deviation and an opposite optimization deviation according to the relative offset data and the opposite offset data; Performing spatial deviation analysis on the identification space data to determine relative deviation directions of the plurality of relative offset identifications and opposite deviation directions of the plurality of opposite offset identifications; According to the relative optimization deviation and the plurality of relative deviation directions, a plurality of corresponding relative offset identifiers are subjected to motion capture prediction optimization processing; According to the opposite optimization deviation and the multiple opposite deviation directions, motion capture prediction optimization processing is performed on multiple corresponding opposite offset identifiers.

7. The method for motion capture prediction and optimization in three-dimensional production according to claim 6, characterized in that: The calculation formula of the relative optimization deviation is: ； in, is the relative optimization deviation, For the The relative distance between the relative offset mark and the corresponding auxiliary motion capture mark is A relative offset marker, is the standard distance; The calculation formula of the opposite optimization deviation is: ； in, To optimize the deviation in opposite directions, For the The relative distance between the opposite offset marks and the corresponding auxiliary motion capture marks is Opposite offset markers.

8. A motion capture prediction and optimization system in three-dimensional production, characterized in that: The system includes a shooting route optimization module, a tracking binocular shooting module, a marker position recognition module, a marker offset recognition module and a motion capture prediction optimization module, wherein: The shooting route optimization module is used to obtain the basic moving route of the action target, plan the basic shooting route, track and monitor the positioning mark of the action target, optimize the basic shooting route in real time, and generate an optimized shooting route; A tracking binocular shooting module is used to track the action target and shoot binocularly according to the optimized shooting route, and obtain binocular shooting data in real time; A marker position recognition module is used to perform position recognition on a plurality of active motion capture markers and corresponding auxiliary motion capture markers of the motion target according to the binocular shooting data, and obtain marker space data; A marker offset identification module, used for performing offset identification on the marker space data, selecting a plurality of relative offset markers and a plurality of opposite offset markers, and extracting relative offset data and opposite offset data; The motion capture prediction optimization module is used to calculate the relative optimization deviation and the opposite optimization deviation according to the relative offset data and the opposite offset data, and perform motion capture prediction optimization processing on multiple active capture identifiers.

9. The motion capture prediction and optimization system in three-dimensional production according to claim 8, characterized in that: The shooting route optimization module specifically includes: A basic acquisition unit, used to acquire the basic moving route and motion capture parameters of the action target; A shooting planning unit, used for performing motion capture shooting planning according to the basic moving route and the motion capture shooting parameters, and generating a basic shooting route; An identification determination unit, used to determine a location identification of the action target; A tracking and monitoring unit, used to track and monitor the positioning mark of the action target during the movement of the action target, and obtain positioning monitoring data; The route optimization unit is used to optimize the basic shooting route in real time according to the positioning monitoring data to generate an optimized shooting route.

10. The motion capture prediction and optimization system in three-dimensional production according to claim 8, characterized in that: The identification offset recognition module specifically includes: A distance recording unit, used for recording relative identification distances between a plurality of active capture identifications and corresponding auxiliary motion capture identifications according to the identification space data; An offset comparison unit, configured to perform an offset comparison on a plurality of relative identification distances based on a preset standard distance and an error tolerance interval, and record an offset comparison result; An identifier selection unit, configured to select a plurality of relative offset identifiers and a plurality of opposite offset identifiers from the plurality of active capture identifiers according to the offset comparison result; The data extraction unit is used to extract relative offset data and opposite offset data from the plurality of relative identification distances.

Images