Clothing 3D model generating device and program

Abstract

To provide a clothing 3D model generating device and a program capable of easily generating clothing 3D model data to meet required quality.SOLUTION: A 3D model generating device 1 comprises a clothing information acquisition part 10 to acquire the clothing information including clothing category information and clothing specification information, a pattern data generating part 20 to generate clothing pattern data based on the clothing information, and a 3D model data generating part 30 of the clothing to generate the clothing 3D model data based on the pattern data. The clothing information acquisition part 10 acquires specification information where the 3D model data generated by the clothing 3D model data generating part 30 meets quality requirements and which is required for the pattern data generation in the pattern data generating part 20.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an apparatus and a program for generating a 3D model of clothing. [Background technology]

[0002] In recent years, many EC (Electronic Commerce) sites selling clothes have appeared. EC sites post many photos of the clothes themselves, photos of how they look on models, and various information about sizes. Consumers refer to this information when purchasing products. However, compared to other categories of products such as electronic appliances, the current situation is that sales of clothes on EC sites are not very active. One of the reasons for this is that consumers have a wide variety of body shapes, making it difficult for consumers to imagine how the clothes will look when they actually wear them. To solve this problem, a virtual try-on system has been proposed. The virtual try-on system prepares a 3D model of an avatar that imitates the consumer's body shape and a 3D model of the clothes, generates an image of the 3D model of the clothes being worn by the 3D model of the avatar, and provides this image to consumers on the EC site. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-170394 A [Patent Document 2] Patent No. 6659901 Summary of the Invention [Problem to be solved by the invention]

[0004] To operate the virtual try-on system described above, it is necessary to prepare 3D model data for the clothing. However, clothing designs are diverse, and the 3D model data differs depending on the design. Furthermore, even if the overall design is the same, the 3D model data differs if the size is different. For this reason, the 3D model data required to operate the virtual try-on system smoothly is large, making it difficult to operate in a practical manner. Therefore, a means of easily obtaining 3D model data for clothing was required.

[0005] One method of acquiring 3D model data for clothing is to generate 3D model data by scanning a model or mannequin wearing the clothing with a 3D scanner. However, the 3D model data generated by this method has wrinkles and shadows on the clothing according to the body shape of the model or mannequin. Therefore, an image in which the 3D model of the clothing is worn by a 3D model of an avatar will differ from the appearance of real clothing worn by a real person. Specifically, the silhouette and texture of the clothing will be significantly different from reality. In addition, since real clothing is required to generate 3D model data using a 3D scanner, it is still not realistic to prepare a large amount of 3D model data.

[0006] Another method of acquiring 3D model data for clothing is to acquire pattern data for the parts that make up the clothing, and generate 3D model data by simulating the assembly (sewing) of the parts in a virtual space (see Patent Document 1). With this method, an image in which a 3D model of clothing is worn by a 3D model of an avatar is closer to the appearance of real clothing worn by a real person, compared to a method using a 3D scanner.

[0007] This pattern data is necessary at the clothing manufacturing site. However, in recent years, with the increase in OEM (Original Equipment Manufacturing) and outsourcing of production, the pattern data is owned by the OEM supplier or the outsourced factory, and even the clothing manufacturer often does not own the pattern data. In particular, OEM suppliers and outsourced factories consider pattern data to be a kind of asset, and tend to avoid providing such assets to outside parties as much as possible. For this reason, it is not realistic for the operator of the virtual fitting system to obtain the pattern data of all products. In addition, since the generation of pattern data relies on the experience and tacit knowledge of people called pattern makers, it is also not realistic for the operator of the virtual fitting system to generate the pattern data himself.

[0008] In addition, the quality required for 3D clothing model data differs depending on various circumstances, such as the purpose of use and the type of clothing. For example, the accuracy of the 3D model data required differs between when it is desired to express the details of the clothing design and when it is sufficient to express a rough silhouette. In addition, the accuracy of the 3D model data required differs between when it is desired to use the 3D clothing model data in a virtual fitting system on an EC site and when it is desired to use the data as clothing data for an avatar in the metaverse. For this reason, there is a demand for easily obtaining 3D clothing model data that meets the required quality.

[0009] The present invention has been made in consideration of the above circumstances, and its object is to provide a 3D clothing model generation device and program that can easily generate 3D clothing model data that meets the required quality. [Means for solving the problem]

[0010] In order to achieve the above object, the present invention provides an apparatus for generating 3D model data for clothing, comprising: clothing information acquisition means for acquiring clothing information including clothing category information and specification information defining the appearance of the clothing; pattern data generation means for generating clothing pattern data based on the clothing information; and clothing 3D model data generation means for generating 3D model data for clothing based on the pattern data, wherein the clothing information acquisition means is configured to ensure that the 3D model data generated by the clothing 3D model data generation means satisfies predetermined quality requirements and acquires specification information required for generating the pattern data in the pattern data generation means. Effect of the Invention

[0011] According to the present invention, the clothing information acquisition means acquires clothing specification information and clothing category required for generating pattern data, the pattern data generation means generates pattern data based on the specification information and category, and the clothing 3D model data generation means generates 3D model data based on the pattern data. Here, the specification information acquired by the clothing information acquisition means is such that the 3D model data generated by the clothing 3D model data generation means satisfies a predetermined quality requirement and is required for generating pattern data in the pattern data generation means. Therefore, it is possible to easily generate 3D model data for clothing that meets the required quality. [Brief description of the drawings]

[0012] [Figure 1] Functional block diagram of a 3D model generating device according to a first embodiment [Diagram 2] An example of a data structure of a category information storage unit [Diagram 3] An example of a data structure of the quality information storage unit [Figure 4] An example of a sales page on an EC site as external data [Diagram 5] An example of processing to extract specification information from an image [Figure 6] Flowchart explaining the operation of the clothing information acquisition unit [Figure 7] Functional block diagram of the virtual fitting device [Figure 8] An example of a generated fitting image [Figure 9] 1 is a functional block diagram of a 3D model generating device according to a second embodiment; [Figure 10] 11 is a functional block diagram of a 3D model generating device according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] (First embodiment) A 3D model generating device according to a first embodiment of the present invention will be described with reference to the drawings. Fig. 1 is a system configuration diagram of the 3D model generating device.

[0014] The 3D model generating device 1 according to the present embodiment acquires category information and specification information of clothing, generates pattern data based on the category information and specification information, and generates 3D model data of clothing based on the pattern data. Here, specification information refers to various information that defines the appearance of clothing. The specification information includes at least dimensions such as length, width, and waist. The specification information can include meta-information other than dimensions, such as the material of the clothing, how the wrinkles are formed, the roundness of the side seams, and other design features. The type and number of specification information required to generate pattern data differs depending on the category of clothing. The type and number of specification information required to generate pattern data also differs depending on the quality classification required for the appearance image of the clothing generated from the 3D model data.

[0015] 1, the 3D model generating device 1 includes a clothing information acquiring unit 10, a pattern data generating unit 20, a clothing 3D model data generating unit 30, a category information storage unit 40, a quality information storage unit 41, a specification information storage unit 50, a pattern data storage unit 60, and a clothing 3D model data storage unit 70. An input device 1a for a user to input various information, and a display device (not shown) are connected to the 3D model generating device 1.

[0016] The 3D model generation device 1 is configured by a well-known information processing device equipped with a main processing unit, a main storage device, an auxiliary storage device, a network interface, and the like. There is no restriction on the implementation form of the 3D model generation device 1. For example, the 3D model generation device 1 may be implemented by distributing each unit across multiple hardware devices. The 3D model generation device 1 may be implemented by installing a program in an information processing device. The 3D model generation device 1 may be implemented as dedicated hardware. The 3D model generation device 1 may be implemented in a cloud server located on the Internet.

[0017] The clothing information acquisition unit 10 acquires data including specification information from outside the 3D model generation device 1 (hereinafter referred to as "external data"), and extracts the specification information necessary for generating pattern data in the subsequent pattern data generation unit 20. The specification information necessary for generating pattern data in the pattern data generation unit 20 differs depending on the clothing category. Therefore, the clothing information acquisition unit 10 performs clothing information acquisition processing based on the category information stored in the category information storage unit 40.

[0018] An example of the data structure of the category information storage unit 40 is shown in FIG. 2. In the example of FIG. 2, the category information includes a category name, a quality classification, and the names of one or more pieces of specification information. Here, a category means a type of clothing such as a T-shirt, pants, a skirt, a shirt, a dress, a jacket, etc. The category can be appropriately set by the user depending on the operating mode, the purpose, etc. In other words, it should be noted that the categories illustrated in FIG. 2 are merely one example, and the present invention can be implemented with other categories.

[0019] The quality category means a category of quality required for the 3D model data generated by the subsequent clothing 3D model data generation unit 30. The required quality category depends on the usage of the 3D model data itself and the usage of the clothing appearance image generated from the 3D model data. In the former case, the required quality category can be set according to, for example, the number of meshes, number of colors, size, etc. of the 3D model data. In the latter case, the required quality category can be set according to the quality of the image generated from the 3D model data. More specifically, it can be set according to the resolution of the image generated from the 3D model data, the resolution, granularity, precision, etc. of the clothing projected on the image.

[0020] In this embodiment, three quality categories, "simple", "detailed" and "metaverse", are set as examples of quality categories. The quality categories "simple" and "detailed" are set according to the purpose of how realistically you want to express the features of the shape and design points of the clothes. For example, when using 3D model data in a fitting system, "simple" is set for fitting when you want to see only the silhouette, and "detailed" is set for fitting when you want to reflect the design in detail. In addition, the quality category "metaverse" is set for when you want to generate 3D model data related to clothes that conforms to the specifications and avatar size of the metaverse space. Note that the quality categories exemplified in FIG. 2 are only examples, and the present invention can be implemented with other quality categories.

[0021] The specification information required for generating paper pattern data in the paper pattern data generating unit 20 differs not only depending on the category but also on the quality classification. Therefore, in this embodiment, the category information storage unit 40 stores specification information for each category and quality classification. If the quality classification is fixed, the quality classification item is not necessary. Also, if the user appropriately updates the category information stored in the category information storage unit 40 according to the quality classification, the quality classification item is not necessary.

[0022] For example, if the category is "T-shirt" and the quality classification is "simple," the specification information is "length, sleeve length, shoulder width, body width." If the category is "T-shirt" and the quality classification is "detailed," the specification information is "material, length, sleeve length, shoulder width, body width, rib width, sleeve width, top width, front drop, back drop, presence and position of pockets, number and position of slits, stitch width." If the category is "T-shirt" and the quality classification is "metaverse," the specification information is "length, sleeve length, shoulder width, body width."

[0023] Furthermore, when the category is "pants" and the quality classification is "simple," the specification information is "rise, inseam, waist, hips." When the category is "pants" and the quality classification is "detailed," the specification information is "material, rise, inseam, waist, hips, hem width, type of pants, number and position of pockets, number and position of belt loops, presence or absence of darts, presence or absence of slits, presence or absence of tucks, stitch width."

[0024] Furthermore, if the category is "skirt" and the quality classification is "simple", the specification information is "skirt length, waist, hips". If the category is "skirt" and the quality classification is "detailed", the specification information is "material, skirt length, waist, hips, hem width, number and position of pleats, number and position of gathers, number and position of slits, stitch width". If the category is "skirt" and the quality classification is "metaverse", the specification information is "skirt length, waist, hips, hem width".

[0025] Furthermore, when the category is "shirt" and the quality classification is "simple", the specification information is "length, width, shoulder width, sleeve length". When the category is "shirt" and the quality classification is "detailed", the specification information is "material, length, width, shoulder width, sleeve length, placket width, top width, front drop, back drop, number and position of tucks, yoke length, presence and position of hemboro, number and position of pockets, collar shape, number and position of buttons, stitch width". When the category is "shirt" and the quality classification is "metaverse", the specification information is "length, width, shoulder width, sleeve length".

[0026] In either category, the difference between the quality classifications "simple" and "detailed" is that the "detailed" quality classification allows the shape of the clothes and the design points of the clothes to be more realistically expressed in the 3D model. For example, when using 3D model data in a fitting system, the quality classification can be set to "simple" for fitting when only the silhouette is to be seen, and "detailed" for fitting when the design is reflected in detail. In addition, when it is desired to generate 3D model data related to clothes according to the specifications of the metaverse space and the avatar size, the quality classification may be generated as "metaverse". Note that the specification information corresponding to each category and quality classification illustrated in FIG. 2 is only one example, and the present invention can be implemented with other specification information.

[0027] The quality information storage unit 45 stores quality information corresponding to the quality classification. This quality information can be used to control the generation process in the pattern data generation process in the pattern data generation unit 20 and to control the output data. In addition, this quality information can be used to control the generation process of 3D model data in the clothing 3D model data generation unit 30 and to control the output data. This quality information can include, for example, size information of the clothing generated in the clothing 3D model data generation unit 30. In addition, the quality information can include, for example, constraint information of the output data in the pattern data generation unit 20 and the clothing 3D model data generation unit 30. For example, for each of the quality classifications "simple", "detailed", and "metaverse", constraint information such as the mesh upper limit number, polygon upper limit number, and data size upper limit of the 3D model data generated in the clothing 3D model data generation unit 30 can be set as quality information as necessary. Furthermore, for example, when the pattern data generated by the pattern data generating unit 20 is in a 3D model format, constraint information such as the upper limit number of meshes, the upper limit number of polygons, and the upper limit of data size of the pattern data generated by the pattern data generating unit 20 can be set as quality information for each of the quality categories "simple", "detailed", and "metaverse" as necessary. Note that the pattern data generating unit 20 and the clothing 3D model data generating unit 30 may perform processing using preset quality information, and only quality information for the quality category to be processed using quality information other than the specified quality information may be stored in the quality information storage unit 45.

[0028] Fig. 3 shows an example of the data structure of the quality information storage unit in this embodiment. Fig. 3 shows an example in which the pattern data generating unit 20 and the clothing 3D model data generating unit 30 perform processing using default quality information for the quality categories "simple" and "detailed", and perform processing using quality information (here, constraint information) stored in the quality information storage unit 45 for the quality category "metaverse". In the example of Fig. 3, for the quality category "metaverse", the size of the clothing generated by the clothing 3D model data generating unit 30 and the upper limit number of meshes are specified. Note that the quality information illustrated in Fig. 3 is merely an example, and the present invention can be implemented with other quality information.

[0029] The clothing information acquisition unit 10 comprises an external data acquisition unit 11, a text information analysis unit 12, an image analysis unit 13, and a user input unit 14. The clothing information acquisition unit 10 acquires external data, and extracts specification information necessary for generating paper pattern data in the subsequent paper pattern data generation unit 20 according to a quality category previously accepted from a user or a default quality category.

[0030] The external data acquisition unit 11 acquires external data including specification information from outside the 3D model generation device 1. The route by which the external data is acquired is not important. For example, the external data acquisition unit 11 acquires external data from another computer 2 via a network. In another example, the external data is acquired from a storage medium 3 connected to the 3D model generation device 1. The data format of the external data is not important. The external data may include characters in text format. The external data may also include an image on which clothing is projected. The external data may also include an image including visualized text. The external data may also include category information together with the specification information. The external data may also be composed of multiple files. For example, the external data may be composed of a group of files written in HTML (Hyper Text Markup Language).

[0031] In this embodiment, the external data acquisition unit 11 can acquire, as external data, a group of files constituting a sales page written in HTML from an EC site 2 that provides a sales page for clothing, via a network. Also, in this embodiment, the external data acquisition unit 11 can acquire external data consisting of one or more files from a locally connected storage medium 3. The source of the external data can be selected based on an instruction from the user.

[0032] FIG. 4 shows an example of a sales page 200 on the EC site 2. As shown in FIG. 4, the sales page 200 includes a clothing image 210 on which the appearance of the clothing is projected, a clothing category 220 described in text format, and clothing dimensions 230 described in text format. Here, the dimensions 230 are the length of the garment. That is, the sales page 200 as external data includes the category and length of the garment described as character information. In the example of FIG. 4, the specification information described as character information is only the length of the garment, but multiple pieces of information may be included. In addition, the length and width of the garment can be calculated by performing image analysis on the clothing image 210. Therefore, the clothing image 210 on the sales page 200 can be interpreted as having dimensional information such as the length and width of the garment potentially embedded therein.

[0033] The character information analysis unit 12 extracts the category and specification information written as characters from the acquired external data. In the example of Fig. 4, the character information analysis unit 12 extracts the category 220 and the length 230 included in the sales page 200.

[0034] The image analysis unit 13 analyzes an image included in the acquired external data, and extracts category and specification information potentially included in the image. For example, when text related to the category and specification information is embedded in the image, the image analysis unit 13 extracts the text using OCR (Optical Character Recognition) technology. In addition, for example, the image analysis unit 13 extracts multiple measurement points of the clothing projected on the image as feature points, and calculates specification information related to dimensions from the distance between the coordinates of the multiple measurement points.

[0035] Further, the image analysis unit 13 presents an image to a user via the user input unit 14, has the user input a plurality of measurement points of the clothing on the image, and calculates the dimension-related specification information based on the coordinates of the inputted plurality of measurement points. FIG. 5 is an example of an image 210 on which a T-shirt is projected. In the example of FIG. 5, the user inputs measurement points 211, 212 on both shoulders of the T-shirt projected on the image 210. The image analysis unit 13 calculates the distance between the coordinates of the measurement points 211, 212 on the image data. The image analysis unit 13 corrects the measured distance to calculate the body width as the specification information in order to match the scale with other specification information related to the calculated already acquired dimensions (body length in the example of FIG. 4). Note that, in the case of an image in which the clothing is worn by a human model, the dimensions serving as the reference in the scale matching correction process can be the dimensions related to the body type of the model reflected in the image (for example, height) in addition to other specification information related to the dimensions already acquired. Also, in the case of an image in which the clothes are worn by a human model or an image in which the clothes are not laid flat, i.e., an image is a projection image in a 3D space, the coordinate system of the 3D space may be recognized, and the distance in the 3D space, i.e., the dimensions of the clothes, may be calculated from the distance between the coordinates of the measurement points 211, 212 on the 2D image data based on the recognition result, and used as the specification information. The coordinate system of the 3D space may be recognized by extracting a plurality of feature points in the image. Note that the specific method of presenting an image to a user, having the user input measurement points, and calculating the dimensions from the input measurement points may be as described in Patent Document 2.

[0036] The user input unit 14 functions as an interface that accepts input of specification information from a user. When accepting the input of specification information, the user input unit 14 can present the name of the specification information to the user. When accepting the input of specification information, the user input unit 14 can present an image included in the external data to the user.

[0037] The operation of the clothing information acquisition unit 10 will be described with reference to the flowchart in Fig. 6. Here, it is assumed that the quality category has been designated in advance by the user.

[0038] The clothing information acquisition unit 10 acquires external data using the external data acquisition unit 11 (step S1). The clothing information acquisition unit 10 extracts a category from the external data using the character information analysis unit 12 (step S2). If the character information analysis unit 12 is unable to extract a category, the clothing information acquisition unit 10 prompts the user to input a category using the user input unit 14, and accepts the category input from the user (steps S3 and S4).

[0039] Next, the clothing information acquisition unit 10 recognizes the specification information required for generating the paper pattern data by referring to the category information stored in the category information storage unit 40 using the acquired category and a pre-specified quality classification as keys, and extracts each piece of specification information from the external data using the character information analysis unit 12 (step S5). If it is not possible to extract all the specification information, the clothing information acquisition unit 10 automatically extracts the remaining specification information from the external data using the image analysis unit 13 (steps S6 and S7).

[0040] If all the specification information required to generate the pattern data cannot be extracted by the above steps, the clothing information acquisition unit 10 will supplement the remaining specification information with input from the user. There are two methods of supplementation: (a) direct input of specification information by the user, and (b) input from the user using an image. The user can freely select which method to use, or whether to use both methods in combination.

[0041] In the case of (a) above, the clothing information acquisition unit 10 prompts the user to input specification information to be completed using the user input unit 14, and accepts the input of specification information from the user (steps S8 and S9).

[0042] In the case of (b) above, the image analysis unit 13 presents an image of the clothing projected onto the user, which is included in the external data, via the user input unit 14, and has the user input a number of measurement points of the clothing onto the image (steps S8, S10 to S11). Next, the image analysis unit 13 measures the distance between the coordinates of the inputted number of measurement points (step S12). Next, the image analysis unit 13 corrects the measured distance to calculate the specification information in order to match the scale with the specification information relating to the dimensions already acquired (step S13).

[0043] Through the above process, all the specification information required to generate the paper pattern data is acquired. The clothing information acquisition unit 10 stores each acquired specification information in the specification information storage unit 50 (step S14). The data format of the specification information storage unit 50 is not important.

[0044] The pattern data generating unit 20 generates pattern data for a garment based on the category, quality classification, and the specification information stored in the specification information storage unit 50, and stores the generated data in the pattern data storage unit 60. Here, when quality information corresponding to the quality classification is stored in the quality information storage unit 45, the pattern data generating unit 20 generates pattern data with a quality corresponding to the quality information. The pattern data generating unit 20 has a different generating unit for each category of garment. In the example of FIG. 1, the pattern data generating unit 20 has a T-shirt generating unit 21 and a pants generating unit 22. The pattern data generating unit 20 generates pattern data by executing the generating units 21 and 22 corresponding to the category. Here, the pattern data includes shape information of a plurality of parts constituting the garment and assembly information of each part. The data format of the pattern data is not important.

[0045] Each of the generation units 21 and 22 for each category is a logic version of the actual pattern creation process performed by a person called a pattern maker. For example, the generation unit 22 for pants first generates data for a belt part based on the specification information. After that, to create the upper part of the body part, the height and width are determined, and the position of the crotch part and the slope of the curve are determined based on the specification information. After that, the lower part is created based on the specification information, and the data for the body part is completed.

[0046] The clothing 3D model data generating unit 30 generates 3D model data of clothing based on the paper pattern data stored in the paper pattern data storage unit 60, and stores the data in the clothing 3D model data storage unit 70. More specifically, the clothing 3D model data generating unit 30 generates 3D model data of clothing by assembling clothing in a virtual space based on the shape data of each part and the assembly information of each part included in the paper pattern data. Here, when quality information corresponding to the quality classification is stored in the quality information storage unit 45, the clothing 3D model data generating unit 30 generates 3D model data with a quality according to the quality information. Any information among the specification information of clothing can be added to the 3D model data of clothing. The specification information added to the 3D model data of clothing includes information that is useful as a parameter when generating an appearance image related to clothing from the 3D model data. For example, meta information other than dimensions such as material. The data format of the 3D model data is not important.

[0047] Next, a virtual try-on device 400 using a 3D model generated by the 3D model generation device 1 will be described with reference to Fig. 7. Here, a user of the 3D model generation device is called a "user of the try-on device" and is distinguished from a user of the 3D model generation device 1. Note that the user of the 3D model generation device 1 and the user of the virtual try-on device 400 may be the same person.

[0048] The virtual try-on device 400 includes a body shape information input section 410, an avatar 3D model generation section 420, a clothing 3D model data storage section 430, and a try-on image generation section 440, as shown in FIG.

[0049] The virtual try-on device 400 is configured by a known information processing device including a main processing unit, a main storage device, an auxiliary storage device, a network interface, and the like. The implementation form of the virtual try-on device 400 is not limited. For example, the virtual try-on device 400 may be implemented by distributing each unit across multiple hardware devices. The virtual try-on device 400 may be implemented by installing a program in an information processing device. The virtual try-on device 400 may be implemented as dedicated hardware. The virtual try-on device 400 may be implemented in a cloud server located on the Internet.

[0050] The body shape information input unit 410 accepts input of body shape information of the user from the user of the virtual fitting device 400. The body shape information may include, for example, dimensional data of each part of a human body such as height, chest circumference, etc. Also, as the body shape information, for example, an image of the user's whole body or a part corresponding to the clothes may be used.

[0051] The avatar 3D model generation unit 420 generates 3D model data of an avatar based on the body shape information input by the body shape information input unit 410. Here, the avatar is an incarnation in the virtual space of the user of the fitting device who is the wearer of the clothes. If the body shape information input by the body shape information input unit 410 is dimensional data, the avatar 3D model generation unit 420 generates 3D model data of an avatar based on the dimensional data. If the body shape information input by the body shape information input unit 410 is image data, the avatar 3D model generation unit 420 analyzes the image data to extract dimensional data of each part of the projected person. Then, the avatar 3D model generation unit 420 generates 3D model data of an avatar based on the extracted dimensional data.

[0052] The clothing 3D model data storage unit 430 stores the clothing 3D model data generated by the 3D model generation device 1. There is no restriction on the method by which the virtual try-on device 400 obtains the 3D model data from the 3D model generation device 1. For example, the 3D model data can be obtained via a network. Also, for example, the 3D model data can be obtained via a specified storage medium.

[0053] The fitting image generating unit 440 generates a wearing image of an avatar wearing clothes based on the 3D model data of the clothes and the 3D model data of the wearer. When the 3D model data of the clothes has specification information added thereto, the fitting image generating unit 440 generates a wearing image taking into account the specification information. For example, when material information is added to the specification information of the 3D model data of the clothes, the fitting image generating unit 440 generates a wearing image with a predetermined texture corresponding to the material. The angle of view, size, etc. of the wearing image are appropriately set by the user of the fitting device. The generated wearing image can be displayed on a predetermined display device (not shown) or stored in a predetermined storage device (not shown). An example of a wearing image 450 generated by the fitting image generating unit 440 is shown in FIG. 8. As shown in FIG. 8, the wearing image 450 shows an avatar 451 projected wearing clothes 452. In particular, the wrinkles and texture of the clothes 451 can be expressed in a manner close to reality.

[0054] As described above, according to the 3D model generating device 1 of this embodiment, the clothing information acquiring unit 10 acquires clothing specification information and clothing category required for generating pattern data, the pattern data generating unit 20 generates pattern data based on the specification information and category, and the clothing 3D model data generating unit 30 generates 3D model data based on the pattern data. Here, the specification information acquired by the clothing information acquiring unit 10 is information required for the 3D model data generated by the clothing 3D model data generating unit 30 to satisfy a predetermined quality requirement and to generate pattern data in the pattern data generating unit 20. Therefore, it is possible to easily generate 3D model data for clothing according to the required quality.

[0055] Furthermore, according to the 3D model generating device 1 of this embodiment, the specification information required to generate the paper pattern data can be automatically acquired from external data. That is, the input work of the specification information required to generate the paper pattern data can be minimized. Therefore, the 3D model data of the clothes can be easily generated.

[0056] Furthermore, according to the 3D model generating device 1 of this embodiment, by setting appropriate category information, it is possible to generate 3D model data of clothing of desired quality. In other words, it is possible to easily generate 3D model data of clothing according to the intended use.

[0057] In this embodiment, the 3D model generation device 1 and the fitting image generation unit 440 are implemented as separate devices, but the functions of the fitting image generation unit 440 may be implemented by being built into the 3D model generation device 1. In this case, it is preferable because the user who creates the 3D model of the clothing can check the finish of the 3D model, etc.

[0058] (Second embodiment) A 3D model generating device according to a second embodiment of the present invention will be described with reference to Fig. 9. Fig. 9 is a system configuration diagram of the 3D model generating device.

[0059] The 3D model generating device 1' according to this embodiment differs from the first embodiment in that it includes a pattern data conversion unit 80. The other configurations are the same as those of the first embodiment, so only the differences will be explained here.

[0060] The pattern data conversion unit 80 converts the dimensions of the pattern data generated by the pattern data generation unit 20. Specifically, the pattern data conversion unit 80 transforms and distorts the clothing pattern data so that it fits the body type of an avatar in a given metaverse. For example, in some metaverses, the body type of an avatar is deformed compared to a real human, such as a three-headed body type. The pattern data conversion unit 80 deforms the pattern data so that it becomes 3D model data of clothing that fits the body type of such an avatar. The pattern data converted by the pattern data conversion unit 80 is stored in the pattern data storage unit 60.

[0061] The 3D model generating device 1' of this embodiment may be configured to be functionally switchable between the 3D model generating device 1 of the first embodiment and the 3D model generating device 1' of this embodiment by switching whether to activate the pattern data conversion unit 80. In this case, as a quality classification in the 3D model generating device 1 of the first embodiment, it is possible to switch whether to activate the pattern data conversion unit 80 according to the quality classification. For example, as a quality classification, "metaverse (deformed)" that is distinguished from other quality classifications may be set. Alternatively, as a category in the 3D model generating device 1 of the first embodiment, a distinction may be made according to not only the type of clothing but also the purpose of use. Taking the category of "T-shirts" as an example, this can be divided into "T-shirts (for normal use)" (where "for normal use" means for use by real people) and "T-shirts (for metaverse)" so that when the category is "T-shirts (for metaverse)", the pattern data conversion unit 80 is activated (for example, switched to the 3D model generation device 1' of this embodiment).

[0062] According to the 3D model generating device of the present embodiment, not only 3D model data of real clothes but also 3D model data of clothes that are deformed from the real clothes can be obtained. Therefore, for example, it can be used in a case where real clothes are sold and clothing data of the same design as the real clothes that can be used in the metaverse is sold.

[0063] (Third embodiment) A 3D model generating device according to a third embodiment of the present invention will be described with reference to Fig. 10. Fig. 10 is a system configuration diagram of the 3D model generating device.

[0064] A 3D model generating device 1'' according to this embodiment differs from the first embodiment in that it includes a 3D model data conversion unit 90. As the other configurations are similar to those of the first embodiment, only the differences will be described here.

[0065] The 3D model data conversion unit 90 performs a conversion process of the dimensions of the 3D model data of the clothes generated by the clothing 3D model data generation unit 30. Specifically, the 3D model data conversion unit 90 transforms and distorts the 3D model data of the clothes so that it matches the body type of the avatar in a certain metaverse. For example, in some metaverses, the body type of the avatar is deformed compared to a real human being, such as a three-head body type. The 3D model data conversion unit 90 deforms the 3D model data of the clothes so that it matches the body type of such an avatar. The 3D model data of the clothes converted by the 3D model data conversion unit 90 is stored in the clothing 3D model data storage unit 70 together with the 3D model data before conversion.

[0066] The 3D model generation device 1" of this embodiment may be configured to be functionally switchable between the 3D model generation device 1 of the first embodiment and the 3D model generation device 1" of this embodiment by switching whether to activate the 3D model data conversion unit 90. In this case, as a quality classification in the 3D model generation device 1 of the first embodiment, it is possible to switch whether to activate the 3D model data conversion unit 90 according to the quality classification. For example, as a quality classification, "metaverse (deformed)" that is distinguished from other quality classifications may be set. Alternatively, as a category in the 3D model generation device 1 of the first embodiment, a distinction may be made not only according to the type of clothing but also according to the purpose of use. Taking the category of "T-shirts" as an example, this can be divided into "T-shirts (normal use)" (where "normal use" means a use that would be worn by a real person) and "T-shirts (for metaverse)" so that when the category is "T-shirts (for metaverse)", the above-mentioned 3D model data conversion unit 90 is activated (for example, switched to the 3D model generation device 1' of this embodiment).

[0067] According to the 3D model generating device of the present embodiment, 3D model data of real clothes and 3D model data of clothes deformed from the real clothes can be obtained. Therefore, for example, it can be used in a case where real clothes are sold and clothing data of the same design as the real clothes that can be used in the metaverse is sold.

[0068] Although the first and second embodiments of the present invention have been described in detail above, the present invention is not limited to the above-mentioned embodiments, and various improvements and modifications may be made without departing from the spirit and scope of the present invention.

[0069] For example, in the above embodiment, external data that directly or potentially includes specification information is acquired from an EC site via a network, but it may be acquired from another site.

[0070] For example, in the above embodiment, three means are provided for extracting specification information from external data: extraction from character information, extraction from an image, and input from a user, but the order in which these means are performed is not important. Also, these means may be performed arbitrarily at the user's discretion.

[0071] In addition, in the above embodiment, the method of acquiring specification information from an image in the image analysis unit 13 requires the user to input the measurement point positions of the clothing, but the measurement points may be set automatically using machine learning or a specified algorithm.

[0072] In addition, in the virtual fitting device 400 in the above embodiment, the user is required to input body shape information such as height and chest circumference, and avatar 3D model data is generated from the input body shape information. However, it is also possible to store a plurality of body shape templates in advance, have the user select which body shape template to use, and generate avatar 3D model data from the selected body shape template. [Explanation of symbols]

[0073] 1. 3D model generation device 10…Clothing information acquisition department 11…External data acquisition section 12...Character information analysis section 13...Image analysis section 14...User input section 20…Pattern data generation section 21…T-shirt printing section 22…Pants forming section 30…Clothing 3D model data generation section 40…Category information storage section 45...Quality information storage section 50…Specification information storage section 60...Pattern data storage section 70…Clothing 3D data storage unit 80...Pattern data conversion section 90...3D model data conversion section 400…Virtual fitting device 410...Body shape information input section 420…Avatar 3D model generation section 430…Clothing 3D model data storage section 440…Fitting image generating unit

Claims

1. 1. An apparatus for generating 3D model data of clothing, comprising: a clothing information acquiring means for acquiring clothing information including clothing category information and specification information defining an appearance of the clothing; a pattern data generating means for generating pattern data for a garment based on the garment information; a clothing 3D model data generating means for generating 3D model data of clothing based on the paper pattern data, The clothing information acquisition means acquires specification information required for the generation of paper pattern data in the paper pattern data generation means, so that the 3D model data generated by the clothing 3D model data generation means satisfies a predetermined quality requirement. A 3D clothing model generating device comprising:

2. The garment information acquisition means acquires part of the specification information required for generating the pattern data in the pattern data generation means from external data, and acquires the remaining specification information required for generating the pattern data in the pattern data generation means from an input by a user.

2. The apparatus for generating a 3D clothing model according to claim 1.

3. The clothing information acquisition means presents names of the remaining specification information to a user and accepts input of the remaining specification information from the user.

3. The apparatus for generating a 3D clothing model according to claim 2.

4. The clothing information acquiring means acquires external data including a character string related to clothing, and extracts first dimensional information as specification information from the character string.

4. The apparatus for generating a 3D clothing model according to claim 1,

5. The clothing information acquiring means acquires external data including an image relating to clothing, and extracts second dimensional information as specification information from the image.

4. The apparatus for generating a 3D clothing model according to claim 1,

6. The clothing information acquisition means acquires external data including character strings and images related to clothing, extracts first dimensional information as specification information from the character strings, extracts second dimensional information as specification information from the images, and corrects the second dimensional information so that the scale of the second dimensional information matches that of the first dimensional information.

4. The apparatus for generating a 3D clothing model according to claim 1,

7. The clothing information acquisition means acquires external data including an image on which clothing is projected, presents the image to a user, and has the user input a plurality of measurement points of the clothing on the image, and calculates second dimensional information based on the coordinates of the input plurality of measurement points.

4. The apparatus for generating a 3D clothing model according to claim 1,

8. The method includes the steps of:

4. The apparatus for generating a 3D clothing model according to claim 1,

9. A body shape information acquisition means for acquiring body shape information of a wearer of the garment; an avatar 3D model generating means for generating 3D model data of an avatar based on the body shape information; and a wearing image generating means for generating a wearing image of an avatar wearing the clothing based on the 3D model data of the clothing and the 3D model data of the avatar.

4. The apparatus for generating a 3D clothing model according to claim 1,

10. A computer is caused to function as each part of the 3D clothing model generating device according to claim 1. A 3D clothing model generation program.

Images