Virtual Reality Processing Measurement Guidance and Verification Method for Remote Fashion Design Customization

Abstract

This invention provides a virtual reality processing measurement guidance and verification method for remote fashion design customization, belonging to the field of fashion design customization technology. The method includes the following steps: constructing a virtual interactive processing system, deployed on a user-operable terminal, possessing core functions related to human body dimensions such as guidance, acquisition, processing, verification, adjustment, and data interaction; pre-setting measurement-related benchmarks and a visual guidance interface; and integrating image processing algorithms and fashion 3D model resources. This invention, through the guidance function of the virtual interactive processing system, solves the problems of non-standard user measurement postures and non-standard use of calibration carriers in existing technologies. The system overlays guidance prompts in real time through a visual guidance interface, clearly informing users of the deployment location of the calibration carrier, acquisition angle, and standard posture requirements, interactively correcting users' non-standard operations, ensuring the standardization and completeness of information acquisition, and providing a foundation for the accuracy of subsequent dimension extraction.

Description

Technical Field

[0001] This invention relates to the field of fashion design customization technology, and in particular to a virtual reality processing measurement guidance and verification method for remote fashion design customization. Background Technology

[0002] With the rapid development of the remote customization industry and the popularization of virtual interaction technology, remote fashion design and customization has gradually become an emerging development direction in the apparel industry. Its core requirement is to achieve accurate acquisition of users' body dimensions and intuitive verification of the fashion customization effect. Currently, remote fashion customization's body measurement methods mainly fall into three categories: First, relying on professional measurement personnel to conduct on-site measurements, obtaining key body dimension parameters through tools such as measuring tapes, and then transmitting them to the design end for customization; second, using professional measurement equipment, where users need to go to a designated location to complete the measurement, and the equipment generates a 3D human body model and dimension data through laser scanning; third, a simple measurement method based on mobile terminals, where users are guided to take human body images through terminal applications, and combined with preset proportions or simple calibration carriers, basic image recognition algorithms are used to extract body dimensions. Some solutions are supplemented with simple virtual try-on functions to achieve a preliminary preview of the customization effect. Meanwhile, the application of existing virtual interaction processing technology in the field of apparel customization is mostly concentrated on virtual try-on displays, rarely integrating it with body measurement guidance and dimension verification to form a complete closed loop, and often relying on specialized equipment or complex operating procedures.

[0003] Specifically, in existing technologies: professional on-site measurement is costly and inefficient, limited by geographical and time constraints, making large-scale remote coverage impossible; professional 3D measurement equipment is expensive, with limited network coverage, resulting in poor user convenience and difficulty in meeting the self-measurement needs of ordinary households; existing simple mobile terminal measurement methods lack standardized interactive guidance, leading to non-standard user shooting postures and improper use of calibration carriers, resulting in poor information collection quality and affecting the accuracy of size extraction; size extraction algorithms are simplistic, lacking coordination with calibration, posture correction, and other processes, resulting in large size errors that cannot meet the precision requirements of fashion customization; the lack of a complete measurement-verification-adjustment closed loop means users cannot intuitively judge whether the extracted dimensions meet their needs, the size adjustment process is cumbersome, and there is a lack of effective means for re-verification after adjustment, making it difficult to guarantee the fit of customized clothing; existing virtual interactive processing technology has low integration with the measurement and verification stages, failing to form a collaborative mechanism of interactive guidance and virtual verification, resulting in a poor user experience. Therefore, this invention proposes a virtual reality processing measurement guidance and verification method for remote fashion design customization to solve the problems existing in the prior art. Summary of the Invention

[0004] To address the aforementioned issues, this invention proposes a virtual reality processing measurement guidance and verification method for remote fashion design customization. Through the guidance function of the virtual interactive processing system, it solves the problems of non-standard user measurement postures and improper use of calibration carriers in existing technologies. The system uses a visual guidance interface to overlay guidance prompts in real time, clearly informing users of the deployment location of the calibration carrier, the collection angle, and the standard posture requirements. It interactively corrects users' non-standard operations, ensuring the standardization and completeness of information collection, providing a foundation for the accuracy of subsequent size extraction. Simultaneously, it simplifies the user operation process and improves the convenience of self-service measurement, allowing even non-professional users to quickly complete standard measurement operations.

[0005] To achieve the objectives of this invention, the invention is implemented through the following technical solution: a virtual reality processing measurement guidance and verification method for remote fashion design customization, comprising the following steps:

[0006] S1: Construct a virtual interactive processing system, deployed on a user-operable terminal, with core functions related to human body size, including guidance, acquisition, processing, verification, adjustment, and data interaction. It also includes preset measurement-related benchmarks, a visual guidance interface, and integrates image processing algorithms and time-lapse 3D model resources.

[0007] S2: The user starts the system. The system guides the user to complete the deployment of the calibration carrier and the collection of human body-related information through a visual guide interface, and simultaneously obtains effective information including human body characteristics and the calibration carrier.

[0008] S3: The system processes effective information through integrated image processing algorithms, extracts key human body size parameters and completes accuracy calibration, and transmits them synchronously to the remote fashion design and customization end.

[0009] S4: Based on the extracted size parameters, the system calls fashion 3D model resources to generate a virtual adaptation model that fits the user's human body characteristics, and displays the adaptation effect through a visual interface to verify the compatibility between size and fashion style.

[0010] S5: Users judge the size fit based on the virtual adaptation effect. If there is any objection, they can input the adjustment command through the system. The system reprocesses the size parameters and updates the virtual adaptation model, repeating the verification process to form a measurement-verification-adjustment collaborative closed loop.

[0011] S6: After the user confirms that there are no errors, the system solidifies the final dimensions and verification information and synchronizes them to the remote design customization terminal, completing the full-process measurement guidance and verification work.

[0012] Further improvements are made in S1, where the calibration carrier for the size is at least one of A4 paper or credit card, and the relevant measurement reference specifications include the actual size parameters of the calibration carrier, the coordinate parameters of the feature points of the calibration carrier, and the measurement posture specifications. The measurement posture specifications include three standard postures: front, side, and back. Each posture presets the placement angle and position parameters of key human body parts - head, shoulders, chest, waist, hips, and limbs. The visual guidance interface overlays posture guidance lines, calibration carrier deployment prompts, and acquisition angle prompts in real time.

[0013] Further improvements are made in the following aspects: In S2, the collection of human body-related information includes image acquisition from three angles: front, side, and back. Two to three images are acquired from each angle. The virtual interactive processing system automatically selects the image with the highest clarity and the most standard posture as valid information. At the same time, the system also has a light compensation function. When the ambient light is insufficient, the brightness of the terminal flash is automatically adjusted to ensure the clarity of the information.

[0014] A further improvement is made in S3, where the image processing algorithm includes a calibration carrier calibration algorithm to complete feature extraction and size calibration of the calibration carrier. The algorithm expression is as follows:

[0015] ,

[0016] Where: U is the actual size parameter of the calibration carrier after calibration, in mm; k is the calibration carrier size conversion factor, which is determined by the ratio of the preset actual size of the calibration carrier to the pixel size in the collected information; P2 is the pixel coordinate of the end feature point in the calibration carrier information, in pixels; P1 is the pixel coordinate of the starting feature point in the calibration carrier information, in pixels; through this algorithm, the pixel size of the calibration carrier in the information is converted into the actual size, providing a standard scale benchmark for subsequent human body size extraction.

[0017] A further improvement is made in S3, where the image processing algorithm includes a human body measurement posture correction algorithm to correct posture deviations during user information collection. The algorithm expression is as follows:

[0018] ,

[0019] in: The deviation angle between the user's actual posture and the preset measurement posture specification, in degrees; The direction vector of key parts of the human body in the user's actual posture is calculated from the coordinates of key human feature points in the collected information. The standard direction vector corresponding to the key parts of the human body in the preset measurement posture specification is a preset fixed value; For vectors with vector The dot product; For vectors The modulus length; For vectors The modulus length; when When the deviation exceeds the preset threshold, the algorithm outputs a correction command and prompts the user to adjust the posture through the system's visual guidance interface until the deviation angle meets the requirements.

[0020] A further improvement is made in S3, where the image processing algorithm includes a human contour extraction algorithm, used to extract complete human contour features from the acquired valid information. The algorithm expression is as follows:

[0021] ,

[0022] Where: Edge(x,y) is the contour edge intensity value of the pixel with coordinates (x,y) in the information, with a value range of [0,255]; (x,y) is the two-dimensional coordinate of the pixel in the information, where x is the horizontal coordinate and y is the vertical coordinate, both in pixels; Gx(x,y) is the gradient value of the pixel in the horizontal direction, calculated by the Sobel operator; Gy(x,y) is the gradient value of the pixel in the vertical direction, calculated by the Sobel operator; max{·} is the maximum value calculation function; when Edge(x,y) is greater than the preset edge threshold, the pixel is determined to be a human contour point, and all contour points are connected in sequence to form complete human contour information for subsequent extraction of key human dimensions.

[0023] A further improvement is made in S3, where the image processing algorithm includes a human body key dimension conversion algorithm, used to calculate the human body key dimension parameters based on the calibrated calibration carrier size and the extracted human body contour. The algorithm expression is as follows:

[0024] ,

[0025] Where: L is the actual size parameter of the key human body parts, in mm; U is the actual size parameter of the calibration carrier after calibration, in mm; Lp is the pixel length of the key human body parts in the information, calculated from the coordinates of the human body contour feature points, in pixels; Up is the pixel length of the calibration carrier in the information, calculated from the coordinates of the calibration carrier feature points, in pixels; through this algorithm, the pixel size of the human body contour is converted into the actual size, realizing the accurate extraction of the key human body dimensions.

[0026] A further improvement is made in S3, where the image processing algorithm includes a size accuracy optimization algorithm to eliminate various errors and improve the accuracy of size extraction. The algorithm expression is as follows:

[0027] ,

[0028] Where: Lopt is the optimized key human body dimension parameter, in mm; L is the key human body dimension parameter initially calculated by the size conversion algorithm, in mm; is the size error correction coefficient, with a value range of [0.1, 0.3], obtained from multiple test calibrations; Ls is the standard reference size of key human body parts, estimated based on basic information such as height and weight input by the user, in mm. This is the attitude deviation correction coefficient, with a value range of [0.05, 0.15], obtained from multiple test calibrations; The residual deviation angle after posture correction is expressed in degrees, which is the deviation angle between the actual posture after correction and the preset measurement posture specification. This algorithm corrects the effects of size conversion errors and residual posture deviations, ensuring that the accuracy of size parameters meets the requirements of fashion customization.

[0029] Further improvements are made in the following: In S4, the virtual adaptation model generation process is as follows: The virtual interactive processing system adjusts the corresponding part size of the preset human virtual model according to the key human body size parameters to make it fit the user's actual body shape, and then attaches the corresponding style model in the fashion 3D model resource to the adjusted human virtual model to generate a virtual adaptation model; the visualization interface supports 360° rotation and scaling of the virtual adaptation model for users to fully view the size adaptation effect.

[0030] Further improvements are made in the following aspects: In S5, the size adjustment command input supports two methods: manually inputting the size adjustment value and selecting the size adjustment level. The adjustment level is preset according to the common size deviations in fashion customization. The data interaction function of the virtual interactive processing system adopts an encrypted transmission method. After receiving the size parameters, the remote fashion design and customization terminal provides real-time feedback on design suggestions, which are then synchronized to the user terminal's visual interface.

[0031] The beneficial effects of this invention are as follows:

[0032] 1. This invention solves the problems of non-standard user measurement posture and non-standard use of calibration carrier in the prior art by using the guidance function of the virtual interactive processing system. The system provides guidance prompts in real time through a visual guidance interface, clearly informing users of the deployment position of the calibration carrier, the collection angle and the standard posture requirements. It interactively corrects users' non-standard operations, ensuring the standardization and completeness of information collection, providing a foundation for the accuracy of subsequent dimension extraction. At the same time, it simplifies the user operation process and improves the convenience of self-service measurement. Even non-professional users can quickly complete standard measurement operations.

[0033] 2. This invention relies on the core functions of a virtual interactive processing system to achieve low-cost self-service measurement, solving the problems of high professional measurement costs and strong dependence on dedicated equipment in existing technologies. Users do not need to equip themselves with professional measurement tools or go to designated service points. They only need to use ordinary calibration carriers and their own user terminals to complete the self-service measurement of human body dimensions through the system. This greatly reduces the measurement cost of remote fashion customization, expands the coverage of remote customization, adapts to the needs of ordinary family users and various consumer groups, and facilitates large-scale promotion and application.

[0034] 3. This invention constructs a complete measurement-verification-adjustment collaborative closed loop through a virtual interactive processing system, solving the problems of lack of a complete verification mechanism and cumbersome size adjustment in existing technologies. The system displays the virtual adaptation effect through a visual interface, intuitively presenting the size adaptation status. Users can view the fit between size parameters and fashion styles in real time. If a problem is found, the user can quickly complete the parameter correction by inputting adjustment commands through the system. After correction, the system reprocesses the size, updates the virtual adaptation display model, and performs re-verification, forming a closed-loop collaboration to ensure that the final size parameters fully meet the user's needs and improve the fit of remote fashion customization.

[0035] 4. The virtual interactive processing system of the present invention integrates multiple sets of collaborative visual processing algorithms, which solves the problems of insufficient size extraction accuracy and single algorithm in the prior art. Multiple sets of dedicated visual processing algorithms respectively realize calibration of the calibration carrier, posture correction, contour extraction, size conversion and accuracy optimization. The algorithms work together to process the collected information in multiple dimensions, effectively eliminating errors caused by collection angle, lighting environment and posture deviation, greatly improving the accuracy of human body size extraction, ensuring that the size parameters meet the precise requirements of fashion customization. At the same time, each algorithm can be adaptively adjusted according to different terminals and different environments, improving the stability of the system.

[0036] 5. This invention achieves deep integration of virtual interactive processing technology with remote fashion customization measurement and verification processes. Through an integrated virtual interactive processing system, it integrates guidance, data acquisition, size extraction, virtual adaptation, size verification, parameter adjustment, and data interaction functions into one system. This solves the problems of low integration and poor user experience in existing virtual interactive technologies, enabling full-process visualization and interactive operation. Users can complete all operations from measurement to verification without switching between multiple platforms or tools. At the same time, the system can be adapted to various mobile terminals, improving the convenience and flexibility of user use and further optimizing the entire process experience of remote fashion customization. Attached Figure Description

[0037] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0038] To enhance understanding of the present invention, the present invention will be further described in detail below with reference to embodiments. These embodiments are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0039] Example 1

[0040] according to Figure 1 As shown, this embodiment proposes a virtual reality processing measurement guidance and verification method for remote fashion design customization, including the following steps:

[0041] The user-operable terminal is a smartphone, and the virtual interactive processing system is in the form of a mobile APP. The calibration media are A4 paper (preset actual size 210mm×297mm) and credit card (preset actual size 85.6mm×53.98mm). Users can choose any calibration media according to their own situation, and the system will automatically match the corresponding benchmark parameters. The preset measurement posture specifications include three postures: front posture (feet together, shoulders level, arms hanging naturally, body upright, face facing the mobile phone camera, lens level with chest), side posture (feet together, body upright, side facing the camera, lens level with waist, arms hanging naturally with close to the body), and back posture (feet together, body upright, back facing the camera, lens level with shoulders, arms hanging naturally).

[0042] The specific design of the visual guidance interface: After launching the mobile APP, the system automatically pops up a visual guidance interface. The left side of the interface displays a schematic diagram of the calibration carrier deployment, and the right side displays the real-time camera image. The image is overlaid with a posture guidance line (a red dotted line, conforming to the human body outline of the preset measurement posture), a calibration carrier deployment prompt box (a blue rectangle, prompting the user to deploy the calibration carrier in the specified position on the screen, ensuring the calibration carrier is fully visible and unobstructed), and a collection angle prompt (text prompt + angle indicator, displaying the deviation between the current collection angle and the standard angle in real time). When the user deploys the calibration carrier, if the calibration carrier is not fully visible or is obstructed, a yellow warning prompt will pop up on the interface, along with a voice reminder: "Please adjust the position of the calibration carrier to ensure it is fully visible and unobstructed." When the user's posture deviates from the preset measurement posture specification, the posture guidance line will turn orange, and a voice reminder will appear: "Please adjust the posture to conform to the guidance line," until the user's posture meets the requirements, the guidance line turns green, and the message "Posture correct, collection can proceed" is displayed.

[0043] The specific information acquisition process is as follows: After the user confirms that the posture and calibration carrier are correctly deployed, they click the acquisition button. The system's information acquisition function automatically acquires images from the corresponding angles, acquiring 3 images for each posture (with an interval of 0.5 seconds). After acquisition, the system automatically performs sharpness analysis (by calculating the variance of image grayscale values; the larger the variance, the higher the sharpness) and posture standardization analysis (by calculating the deviation angle through a posture correction algorithm) on the 3 images. The image with the highest sharpness (grayscale value variance ≥ 800) and the smallest posture deviation angle (θ ≤ 3°) is selected as the valid information for subsequent algorithm processing. If the ambient light is insufficient (average image grayscale value ≤ 120), the system automatically turns on the phone's flash and adjusts the flash brightness to a suitable level (automatically adjusted according to the ambient light intensity, divided into low, medium, and high levels) to ensure that the acquired information is clearly discernible and to avoid contour extraction errors caused by insufficient light. In this embodiment, through the system's guidance function, the average user acquisition time does not exceed 5 minutes, and the success rate of valid information acquisition reaches 99%. Compared with the existing unguided acquisition method, posture standardization is improved by 85%, and information sharpness is improved by 70%.

[0044] Example 2

[0045] according to Figure 1 As shown, this embodiment proposes a virtual reality processing measurement guidance and verification method for remote fashion design customization, including the following steps:

[0046] The specific implementation of the calibration carrier calibration algorithm: After the effective information is transmitted to the system's processing module, the calibration carrier calibration algorithm is first activated to extract features and calibrate the size of the calibration carrier in the information. Taking A4 paper as the calibration carrier as an example, the algorithm first performs image grayscale processing (converting the color image to a grayscale image with a grayscale value range of [0,255]), and then uses a threshold segmentation algorithm (with a preset threshold of 150; pixels with a grayscale value greater than 150 are determined as background, and pixels with a grayscale value less than or equal to 150 are determined as the calibration carrier area) to segment out the calibration carrier area and eliminate background interference. Subsequently, the corner detection algorithm (Harris corner detection) is used to extract the four vertices of the calibration carrier as feature points, and the pixel coordinates of each feature point are determined as P1 (top left vertex), P2 (top right vertex), P3 (bottom left vertex), and P4 (bottom right vertex).

[0047] The specific calculation process of the calibration carrier calibration algorithm: Taking the calibration of the length of A4 paper (297mm) as an example, the pixel coordinates of P1 are (120, 350), and the pixel coordinates of P2 are (880, 352). Then, the pixel length Up of the calibration carrier in the information is Up = P2 - P1 = 880 - 120 = 760 pixels (since the y coordinates of P1 and P2 are similar, the difference in y coordinates is ignored, and the difference in x coordinates is taken as the pixel length). The preset actual size of the calibration carrier is U = 297mm, so the size conversion factor k = U / Up = 297 / 760 ≈ 0.391mm / pixel. This factor will be used as the standard scale benchmark for subsequent human body size conversion. If a credit card is used as the calibration carrier, and the length of the credit card is 85.6 mm, assuming the extracted feature points P1(200,400) and P2(410,402), then Up = 410 - 200 = 210 pixels, and k = 85.6 / 210 ≈ 0.408 mm / pixel. In this embodiment, the calibration error of the calibration carrier calibration algorithm is ≤0.1 mm, ensuring the accuracy of the scale reference.

[0048] The specific implementation of the human contour extraction algorithm: After the calibration carrier is calibrated, the human contour extraction algorithm is started to extract the human contour from the collected valid information. First, the valid information is preprocessed, including Gaussian blurring (Gaussian kernel size is 3×3, standard deviation σ=1.0) to eliminate information noise and avoid incorrect contour extraction caused by noise; then, the horizontal gradient Gx and vertical gradient Gy of each pixel in the information are calculated by the edge detection algorithm (Sobel operator), and then the contour edge intensity value Edge(x,y) is calculated.

[0049] The specific calculation process of the human contour extraction algorithm is as follows: Taking a pixel (x=300, y=500) in the information as an example, the Sobel operator calculates Gx=25 and Gy=18, so Edge(x,y)=|25|+|18|=43. The preset edge threshold is 30. Since 43>30, this pixel is determined to be a human contour point. If a pixel (x=400, y=600) has Gx=10 and Gy=12, then Edge(x,y)=22<30, and this pixel is determined to be a non-contour point. All pixels in the information are traversed sequentially, and all contour points are filtered out. Then, morphological processing (dilation + erosion) is used to fill in small gaps in the contour, eliminate isolated contour points, and form complete human contour information for subsequent size extraction. In this embodiment, the human contour extraction algorithm achieves a contour integrity of 98.5%, accurately extracting the contours of key parts of the human body such as the head, shoulders, chest, waist, hips, and limbs. The contour extraction error is ≤1 pixel, providing a precise contour foundation for subsequent size extraction. The two algorithms work together to ensure the accuracy of the scale reference and contour information, providing double assurance for the accuracy of size extraction.

[0050] Example 3

[0051] according to Figure 1 As shown, this embodiment proposes a virtual reality processing measurement guidance and verification method for remote fashion design customization, including the following steps:

[0052] The specific implementation of the posture correction algorithm: After the human body contour is extracted, the posture correction algorithm is activated to correct posture deviations during the user information collection process. Taking a frontal posture as an example, the standard direction vector of the human shoulder in the preset posture measurement specification is used. The default value is (1,0) (horizontal direction). Using a human contour extraction algorithm, two feature points (left shoulder vertex and right shoulder vertex) are extracted from the human shoulder area. The direction vector of the shoulder in the user's actual pose is then calculated. Assuming the left shoulder vertex pixel coordinates are (250, 420) and the right shoulder vertex pixel coordinates are (750, 425), then the vector... The coordinates are (750-250, 425-420) = (500, 5), vector with vector dot product =500×1+5×0=500, vector Length of the module =√(500 2 +5 2 )≈500.025, vector Length of the module =1, then the deviation angle θ=arccos(500 / (500.025×1))≈0.57°, which is less than the preset deviation threshold of 3°, so the posture is considered qualified and no further correction is needed. If the user's shoulder is tilted, assuming the pixel coordinates of the right shoulder vertex are (740,450), then the vector =(740-250,450-420)=(490,30), dot product=490×1+30×0=490, =√(490²+30²)≈490.938, θ=arccos(490 / 490.938)≈3.45°, which is greater than the preset threshold. The algorithm outputs a correction instruction and prompts the user through the system's visual guidance interface to "Please adjust your shoulder posture and keep your shoulders level" until the deviation angle is ≤3°.

[0053] The specific implementation of the human body size conversion algorithm: After the posture correction is qualified, the size conversion algorithm is started. Based on the calibrated calibration carrier size and the extracted human body contour, the key human body size parameters are converted. Taking shoulder width as an example, the pixel coordinates of the left and right shoulder vertices are extracted through the human body contour extraction algorithm, and the pixel length of the shoulder width Lp=750-250=500 pixels (ignoring the difference in y coordinates); the calibration carrier calibration algorithm yields Up=760 pixels and U=297mm, so the actual shoulder size L=(U×Lp) / Up=(297×500) / 760≈194.08mm. Taking waist circumference as an example, the leftmost and rightmost feature points of the waist contour are extracted, with pixel coordinates of (300,650) and (700,652) respectively, so Lp=700-300=400 pixels, L=(297×400) / 760≈155.26mm. In this embodiment, the initial conversion error of the size conversion algorithm is ≤0.3mm.

[0054] The specific implementation of the dimensional accuracy optimization algorithm: After the dimensional conversion is completed, the dimensional accuracy optimization algorithm is activated to correct various errors. The preset values ​​are α=0.2 (dimensional error correction coefficient) and β=0.1 (posture deviation correction coefficient). The user inputs a height of 175cm and a weight of 65kg. Based on the preset standard reference size table, the estimated standard reference size for the shoulder is Ls=193mm, and the standard reference size for the waist is Ls=155mm. The residual deviation angle after posture correction is Δθ=0.57° (shoulder deviation angle). The optimized shoulder dimension Lopt = 194.08 + 0.2 × (193 - 194.08) + 0.1 × 0.57 = 194.08 - 0.216 + 0.057 ≈ 193.92 mm; the optimized waist dimension Lopt = 155.26 + 0.2 × (155 - 155.26) + 0.1 × 0.57 = 155.26 - 0.052 + 0.057 ≈ 155.27 mm. In this embodiment, through the collaborative processing of three algorithms, the accuracy of extracting key human body dimensions can reach ±0.5 mm, fully meeting the accuracy requirements of remote fashion customization (customization accuracy requirement ±1 mm).

[0055] Example 4

[0056] according to Figure 1 As shown, this embodiment proposes a virtual reality processing measurement guidance and verification method for remote fashion design customization, including the following steps:

[0057] The virtual adaptation model generation process: After dimensional accuracy optimization, the system's processing module transmits the optimized key human body dimensional parameters (shoulder width 193.92mm, waist 155.27mm, chest 100.5mm, hip 105.3mm, sleeve length 58.2mm, etc.) to the virtual verification module. The virtual verification module has a pre-set virtual human body base model (containing basic templates for different heights and body types). Based on the received dimensional parameters, it automatically adjusts the corresponding dimensions of the virtual human body base model: adjusting the shoulder width to 193.92mm, waist 155.27mm, chest 100.5mm, hip 105.3mm, and sleeve length to 58.2mm, generating a personalized virtual human body model that fits the user's actual body shape. Subsequently, the user selects the desired fashion style (such as a suit or dress) in the APP. The virtual verification module calls the corresponding 3D model from the system's preset fashion 3D model resources. This model has preset adaptation parameters for different sizes. The system automatically adjusts the size of the shoulders, waist, chest, and other parts of the fashion 3D model according to the size of the user's personalized human virtual model, so that it fits the human virtual model and generates a virtual adaptation model.

[0058] The specific method of virtual adaptation verification: After the virtual adaptation model is generated, it is displayed intuitively through the system's visual interface. The interface supports 360° rotation and zooming of the virtual adaptation model (users can rotate the model by swiping the screen, and zoom in or out with two fingers), making it easy for users to fully view the size adaptation effect. At the same time, the interface displays the corresponding key human body size parameters. Users can click on any size parameter to view the adaptation details of the corresponding clothing part (e.g., clicking on shoulder width highlights the shoulder area and prompts "Shoulder size fits, no deviation" or "Shoulder size is too wide, adjustment is recommended"). If the virtual adaptation model has size adaptation problems (such as the waist being too tight or the sleeve length being too long), the interface will pop up a red prompt box, indicating the problematic part and the deviation value, and simultaneously issue a voice reminder "There is a size adaptation problem, please adjust the size parameters."

[0059] Size adjustment process and closed-loop implementation: After viewing the virtual fitting effect, if the user finds the waist size too tight (deviation 0.5mm), they can input an adjustment command through the system's parameter adjustment function. The adjustment method is manual input; enter "+0.5mm" in the waist size input box and click confirm. The system transmits the adjustment command to the processing module, which activates the size accuracy optimization algorithm to recalculate the waist size based on the command: L=155.26+0.5=155.76mm, Lopt=155.76+0.2×(155-155.76)+0.1×0.57=155.76-0.152+0.057≈155.66mm. The optimized waist size is then transmitted to the virtual verification module, which synchronously updates the virtual fitting model, adjusts the waist size to 155.66mm, generates a new virtual fitting model, and displays it to the user. The user reviews the virtual fit again to confirm the waist size fits correctly, completing the adjustment loop. If the user still has objections, the adjustment process can be repeated until the size fits correctly. In this embodiment, the size adjustment response time is ≤1 second, the average time to complete the adjustment loop is no more than 1 minute, and the size adjustment success rate reaches 98.5%. Compared with the existing non-loop adjustment method, user satisfaction is improved by 80%.

[0060] Example 5

[0061] according to Figure 1 As shown, this embodiment proposes a virtual reality processing measurement guidance and verification method for remote fashion design customization, including the following steps:

[0062] Cross-terminal adaptation implementation: The virtual interactive processing system of the present invention supports multiple types of user-operable terminals, including smartphones (iOS system, Android system) and tablet computers (iOS system, Android system). The system automatically adjusts the layout of the visual guidance interface, the size of the guide lines, and the parameters of information collection according to the screen size, resolution, and camera parameters of different terminals, so as to ensure that stable and accurate measurement guidance and verification functions can be achieved on different terminals. Taking smartphones (6.7-inch screen, 2400×1080 resolution) and tablets (11-inch screen, 2560×1600 resolution) as examples, the system automatically identifies the terminal type and adjusts the size of the calibration carrier prompt box in the visual guidance interface (60% of the screen width on mobile devices, 50% on tablets), the thickness of the posture guide line (2px on mobile devices, 3px on tablets), and the font size of the acquisition angle prompt (14-point font on mobile devices, 16-point font on tablets). Simultaneously, based on the pixel parameters of the terminal's camera (50MP on mobile devices, 40MP on tablets), the system adjusts the information acquisition resolution (4000×3000 on mobile devices, 3840×2160 on tablets) to ensure information acquisition quality. Furthermore, the system supports user account synchronization. After a user completes a measurement on one terminal, information such as size parameters and virtual adaptation records will be synchronized to the cloud. Users can directly view this information after switching to another terminal and logging in, eliminating the need for repeated measurements and improving user flexibility.

[0063] The specific implementation of encrypted data transmission: The system's data interaction function employs the AES-256 encryption algorithm to encrypt users' body size parameters, personal information (name, contact information, height, weight, etc.), virtual adaptation records, and other information before transmitting it to the cloud server and remote fashion design customization terminal via HTTPS protocol. This ensures the security of data transmission and prevents data leakage and tampering. The specific process is as follows: After the user completes size verification, the system encrypts the final size parameters and confirmation information (the encryption key is negotiated and generated between the user terminal and the cloud server, and the key changes randomly with each transmission). The encrypted data is transmitted to the cloud server, which decrypts and stores the information, while simultaneously forwarding the encrypted data to the remote fashion design customization terminal. Upon receiving the data, the remote fashion design customization terminal decrypts it using the corresponding decryption key to obtain the user's size parameters and confirmation information for subsequent fashion design customization. During data transmission, the system monitors the transmission status in real time. If data loss, tampering, or other abnormalities occur, the system automatically retransmits the data and issues an alert, ensuring the integrity and security of data transmission.

[0064] Collaborative Interaction with Remote Fashion Design and Customization: The remote fashion design and customization platform is equipped with a corresponding receiving terminal. After receiving the size parameters and virtual adaptation confirmation information transmitted by the user, designers can view the user's body size parameters and virtual adaptation effect. Based on the fashion design requirements, they can professionally evaluate the size parameters. If any unreasonable size parameters are found (such as excessive deviation in a certain part of the size, affecting the fashion pattern), design suggestions can be sent through the receiving terminal (e.g., "It is recommended to adjust the waist circumference to 156mm to improve the fit of the fashion pattern"). The design suggestions are encrypted before being transmitted to the user terminal. After receiving the suggestions, the user terminal displays them through the system's visual interface. The user can adjust the size parameters according to the suggestions, and after adjustment, retransmit the confirmation information, forming a collaborative interaction mechanism of "user measurement - designer evaluation - user adjustment - designer confirmation". In this embodiment, the data transmission delay is ≤2 seconds, the encrypted transmission success rate reaches 100%, the cross-terminal adaptation rate reaches 100%, and the collaborative interaction efficiency between the design end and the user end is improved by 75%, significantly shortening the cycle of remote fashion design and customization.

[0065] Validation data:

[0066] To verify the practicality, accuracy, and superiority of this invention, 500 user tests were conducted (covering users of different ages, heights, and body types, including 250 male and 250 female groups). A4 paper and credit cards were used as two common calibration media, and tests were performed on smartphones and tablets. The results were compared with existing remote measurement methods. Test indicators included measurement accuracy, measurement time, successful acquisition of effective information, success rate of size adjustment, user satisfaction, and measurement cost. Specific test data are as follows:

[0067] Measurement Accuracy: The measurement accuracy of key human body dimensions in this invention can reach ±0.5mm, with an average measurement error of 0.32mm for core dimensions such as shoulder, waist, and chest. Existing professional on-site measurements have an average error of 0.4mm, existing simplified mobile terminal measurement methods have an average error of 1.8mm, and existing 3D scanner measurement methods have an average error of 0.35mm. The measurement accuracy of this invention is close to that of a professional 3D scanner and superior to existing simplified mobile terminal measurement methods, fully meeting the accuracy requirements (±1mm) for remote fashion customization.

[0068] Measurement time: The average time for user-initiated self-measurement using this invention is 4.8 minutes (including guidance, data acquisition, dimension extraction, and virtual verification); the average time for on-site measurement by existing professionals is 30 minutes (including on-site visit, measurement, and information recording); the average time for existing simplified mobile terminal measurement methods is 8.5 minutes (including data acquisition and dimension extraction); and the average time for existing 3D scanner measurements is 12 minutes (including queuing, measurement, and information export). The method of this invention significantly reduces measurement time, improving the efficiency of remote measurement.

[0069] Effective information collection success rate: Through the guidance function of the virtual interactive processing system, the present invention achieves an effective information collection success rate of 99%, while the existing mobile terminal measurement method without guidance has an effective information collection success rate of 72%. The main difference is that the guidance mechanism of the present invention can effectively correct users' non-standard operations and ensure the quality of information collection.

[0070] Size adjustment success rate: The present invention constructs a complete measurement-verification-adjustment collaborative closed loop through a virtual interactive processing system, achieving a size adjustment success rate of 98.5% and a user-adjusted size fit of 99.2%. Existing measurement methods without closed-loop adjustment have a size adjustment success rate of 75% and a post-adjustment size fit of 80%. The closed-loop mechanism of the present invention significantly improves the effectiveness and fit of size adjustment.

[0071] User satisfaction: User feedback was collected through a questionnaire survey. The user satisfaction rate of the method of this invention reached 96%, with 97% satisfaction with the ease of system guidance, 95% satisfaction with measurement accuracy, 98% satisfaction with virtual adaptation verification, and 96% satisfaction with the ease of size adjustment. The average user satisfaction rate of existing remote measurement methods is 70%, and the user experience of the method of this invention is significantly better than that of existing technologies.

[0072] Measurement Cost: This invention requires no specialized equipment or personnel, only a standard calibration carrier and the user's own terminal. Measurement can be completed through a virtual interactive processing system, with a single measurement cost approaching zero. Existing professional on-site measurements cost approximately 50 yuan per measurement; existing 3D scanner measurements cost approximately 30 yuan per measurement; while existing simple mobile terminal measurement methods require no additional cost, their accuracy is insufficient, necessitating multiple adjustments, indirectly increasing customization costs. This invention significantly reduces the measurement cost of remote fashion customization, facilitating large-scale application.

[0073] In summary, this invention addresses many shortcomings of existing remote fashion customization measurement and verification technologies, and has significant advantages in terms of measurement accuracy, convenience, efficiency, cost, and user experience. It can effectively support the precise and convenient implementation of remote fashion design and customization.

[0074] This invention addresses the issues of non-standard user measurement postures and improper use of calibration carriers in existing technologies through the guidance function of a virtual interactive processing system. The system provides real-time guidance prompts via a visual interface, clearly informing users of the calibration carrier's deployment location, acquisition angle, and standard posture requirements. It interactively corrects users' non-standard operations, ensuring the standardization and completeness of information collection and laying the foundation for accurate subsequent dimensional extraction. Simultaneously, it simplifies the user operation process and enhances the convenience of self-service measurement, allowing even non-professional users to quickly complete standard measurement operations. Furthermore, relying on the core functions of the virtual interactive processing system, this invention achieves low-cost self-service measurement, solving the problems of high costs and strong reliance on specialized equipment in existing technologies. Users do not need to equip themselves with professional measuring tools or travel to designated locations; they only need to use ordinary calibration carriers and their own user terminals to complete self-service measurement of human body dimensions through the system. This significantly reduces the measurement costs of remote fashion customization, expands the coverage of remote customization, and caters to the needs of ordinary households and various consumer groups, facilitating large-scale promotion and application. Meanwhile, this invention constructs a complete measurement-verification-adjustment collaborative closed loop through a virtual interactive processing system, solving the problems of lacking a complete verification mechanism and cumbersome size adjustment in existing technologies. The system displays the virtual adaptation effect through a visual interface, intuitively presenting the size adaptation status. Users can view the fit between size parameters and fashion styles in real time. If problems are found, they can quickly complete parameter correction by inputting adjustment commands through the system. After correction, the system reprocesses the size, updates the virtual adaptation display model, and performs re-verification, forming a closed-loop collaboration to ensure that the final size parameters fully meet user needs and improve the fit of remote fashion customization. In addition, the virtual interactive processing system of this invention integrates multiple sets of collaborative visual processing algorithms, solving the problems of insufficient size extraction accuracy and single algorithm in existing technologies. Multiple sets of dedicated visual processing algorithms respectively realize calibration of the calibration carrier, posture correction, contour extraction, size conversion, and accuracy optimization. The algorithms work together to process the collected information in multiple dimensions, effectively eliminating errors caused by collection angle, lighting environment, and posture deviation, greatly improving the accuracy of human body size extraction, ensuring that the size parameters meet the precise requirements of fashion customization. At the same time, each algorithm can adaptively adjust according to different terminals and different environments, improving the stability of the system. Finally, this invention achieves deep integration of virtual interactive processing technology with remote fashion customization measurement and verification processes. Through an integrated virtual interactive processing system, it integrates guidance, data acquisition, size extraction, virtual adaptation, size verification, parameter adjustment, and data interaction functions into one system. This solves the problems of low integration of virtual interactive technology and poor user experience in existing technologies, and realizes full-process visualization and interactive operation. Users can complete all operations from measurement to verification without switching between multiple platforms or tools. At the same time, the system can be adapted to various mobile terminals, improving the convenience and flexibility of user use, and further optimizing the entire process experience of remote fashion customization.

[0075] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A virtual reality processing measurement guidance and verification method for remote fashion design customization, characterized in that, Includes the following steps: S1: Construct a virtual interactive processing system, deployed on a user-operable terminal, with core functions related to human body size, including guidance, acquisition, processing, verification, adjustment, and data interaction. It also includes preset measurement-related benchmarks, a visual guidance interface, and integrates image processing algorithms and time-lapse 3D model resources. S2: The user starts the system. The system guides the user to complete the deployment of the calibration carrier and the collection of human body-related information through a visual guide interface, and simultaneously obtains effective information including human body characteristics and the calibration carrier. S3: The system processes effective information through integrated image processing algorithms, extracts key human body size parameters and completes accuracy calibration, and transmits them synchronously to the remote fashion design and customization end. S4: Based on the extracted size parameters, the system calls fashion 3D model resources to generate a virtual adaptation model that fits the user's human body characteristics, and displays the adaptation effect through a visual interface to verify the compatibility between size and fashion style. S5: Users judge the size fit based on the virtual adaptation effect. If there is any objection, they can input the adjustment command through the system. The system reprocesses the size parameters and updates the virtual adaptation model, repeating the verification process to form a measurement-verification-adjustment collaborative closed loop. S6: After the user confirms that there are no errors, the system solidifies the final dimensions and verification information and synchronizes them to the remote design customization terminal, completing the full-process measurement guidance and verification work.

2. The virtual reality processing measurement guidance and verification method for remote fashion design customization according to claim 1, characterized in that: In S1, the calibration carrier for the size is at least one of A4 paper and credit card. The relevant measurement reference specifications include the actual size parameters of the calibration carrier, the coordinate parameters of the feature points of the calibration carrier, and the measurement posture specifications. The measurement posture specifications include three standard postures: front, side, and back. Each posture presets the placement angle and position parameters of key human body parts - head, shoulders, chest, waist, hips, and limbs. The visual guidance interface overlays posture guidance lines, calibration carrier deployment prompts, and acquisition angle prompts in real time.

3. The virtual reality processing measurement guidance and verification method for remote fashion design customization according to claim 1, characterized in that: In S2, the collection of human body-related information includes image acquisition from three angles: front, side, and back. Two to three images are acquired from each angle. The virtual interactive processing system automatically selects the image with the highest clarity and the most standard posture as valid information. At the same time, the system also has a light compensation function. When the ambient light is insufficient, the brightness of the terminal flash is automatically adjusted to ensure the clarity of the information.

4. The virtual reality processing measurement guidance and verification method for remote fashion design customization according to claim 1, characterized in that: In step S3, the image processing algorithm includes a calibration carrier calibration algorithm, which is used to complete the feature extraction and size calibration of the calibration carrier. The algorithm expression is as follows: ， Where: U is the actual size parameter of the calibration carrier after calibration, in mm; k is the calibration carrier size conversion factor, which is determined by the ratio of the preset actual size of the calibration carrier to the pixel size in the collected information; P2 is the pixel coordinate of the end feature point in the calibration carrier information, in pixels; P1 is the pixel coordinate of the starting feature point in the calibration carrier information, in pixels; through this algorithm, the pixel size of the calibration carrier in the information is converted into the actual size, providing a standard scale benchmark for subsequent human body size extraction.

5. The virtual reality processing measurement guidance and verification method for remote fashion design customization according to claim 4, characterized in that: In step S3, the image processing algorithm includes a human body measurement posture correction algorithm, used to correct posture deviations during user information collection. The algorithm expression is as follows: ， in: The deviation angle between the user's actual posture and the preset measurement posture specification, in degrees; The direction vector of key parts of the human body in the user's actual posture is calculated from the coordinates of key human feature points in the collected information. The standard direction vector corresponding to the key parts of the human body in the preset measurement posture specification is a preset fixed value; For vectors with vector The dot product; For vectors The modulus length; For vectors The modulus length; when When the deviation exceeds the preset threshold, the algorithm outputs a correction command and prompts the user to adjust the posture through the system's visual guidance interface until the deviation angle meets the requirements.

6. The virtual reality processing measurement guidance and verification method for remote fashion design customization according to claim 5, characterized in that: In step S3, the image processing algorithm includes a human contour extraction algorithm, which is used to extract complete human contour features from the acquired valid information. The algorithm expression is as follows: ， Where: Edge(x,y) is the contour edge intensity value of the pixel with coordinates (x,y) in the information, with a value range of [0,255]; (x,y) is the two-dimensional coordinate of the pixel in the information, where x is the horizontal coordinate and y is the vertical coordinate, both in pixels; Gx(x,y) is the gradient value of the pixel in the horizontal direction, calculated by the Sobel operator; Gy(x,y) is the gradient value of the pixel in the vertical direction, calculated by the Sobel operator; max{·} is the maximum value calculation function; when Edge(x,y) is greater than the preset edge threshold, the pixel is determined to be a human contour point, and all contour points are connected in sequence to form complete human contour information for subsequent extraction of key human dimensions.

7. The virtual reality processing measurement guidance and verification method for remote fashion design customization according to claim 6, characterized in that: In step S3, the image processing algorithm includes a human body key dimension conversion algorithm, which is used to calculate the human body key dimension parameters based on the calibrated calibration carrier size and the extracted human body contour. The algorithm expression is as follows: ， Where: L is the actual size parameter of the key human body parts, in mm; U is the actual size parameter of the calibration carrier after calibration, in mm; Lp is the pixel length of the key human body parts in the information, calculated from the coordinates of the human body contour feature points, in pixels; Up is the pixel length of the calibration carrier in the information, calculated from the coordinates of the calibration carrier feature points, in pixels; through this algorithm, the pixel size of the human body contour is converted into the actual size, realizing the accurate extraction of the key human body dimensions.

8. The virtual reality processing measurement guidance and verification method for remote fashion design customization according to claim 7, characterized in that: In step S3, the image processing algorithm includes a size accuracy optimization algorithm to eliminate various errors and improve the accuracy of size extraction. The algorithm expression is as follows: ， Where: Lopt is the optimized key human body dimension parameter, in mm; L is the key human body dimension parameter initially calculated by the size conversion algorithm, in mm; is the size error correction coefficient, with a value range of [0.1, 0.3], obtained from multiple test calibrations; Ls is the standard reference size of key human body parts, estimated based on basic information such as height and weight input by the user, in mm. This is the attitude deviation correction coefficient, with a value range of [0.05, 0.15], obtained from multiple test calibrations; The residual deviation angle after posture correction is expressed in degrees, which is the deviation angle between the actual posture after correction and the preset measurement posture specification. This algorithm corrects the effects of size conversion errors and residual posture deviations, ensuring that the accuracy of size parameters meets the requirements of fashion customization.

9. The virtual reality processing measurement guidance and verification method for remote fashion design customization according to claim 1, characterized in that: In step S4, the virtual adaptation model generation process is as follows: the virtual interactive processing system adjusts the corresponding part size of the preset human virtual model according to the key human body size parameters to make it fit the user's actual body shape, and then attaches the corresponding style model in the fashion 3D model resource to the adjusted human virtual model to generate the virtual adaptation model; the visualization interface supports 360° rotation and scaling of the virtual adaptation model for users to fully view the size adaptation effect.

10. The virtual reality processing measurement guidance and verification method for remote fashion design customization according to claim 1, characterized in that: In S5, the size adjustment command input supports two methods: manually inputting the size adjustment value and selecting the size adjustment level. The adjustment level is preset according to the common size deviations in fashion customization. The data interaction function of the virtual interactive processing system adopts encrypted transmission. After receiving the size parameters, the remote fashion design and customization terminal provides real-time feedback on design suggestions, which are synchronized to the user terminal's visual interface.

Images