WO2015052476A1

WO2015052476A1 - Method for generating body measurement data of a user and system for selecting a set of articles of clothing for a user

Info

Publication number: WO2015052476A1
Application number: PCT/GB2014/051923
Authority: WO
Inventors: Daniel Laurence
Original assignee: Isizeme Ltd.
Priority date: 2013-10-07
Filing date: 2014-06-24
Publication date: 2015-04-16
Also published as: GB201411190D0; GB201317658D0; GB2518931A

Abstract

A method for generating body measurement data of a user, the method comprising generating image data representing at least one image of the user, the or each image including a calibration portion with at least one known dimension, generating a user representation including multiple user body measurements using the image data, and corresponding system.

Description

METHOD FOR GENERATING BODY MEASUREMENT DATA OF A USER AND SYSTEM FOR SELECTING A SET OF ARTICLES OF CLOTHING FOR A USER

TECHNICAL FIELD

Aspects relate, in general, to a method for generating body measurement data of a user and system for selecting a set of articles of clothing for a user.

BACKGROUND

Internet shopping is widespread and affords advantages for both consumer and retailer. For example, in terms of clothes and fashion shopping, a consumer is able to browse images of large numbers of articles of clothing online, often far more quickly than would be possible compared to physically browsing through articles in a store. The consumer can also typically purchase online without having to travel to a retail outlet. From a retailer perspective, offering online shopping can enhance brand presence and reduce overheads, as it may be possible to reduce retail space and associated staffing for example.

However, there are several drawbacks of online clothes shopping. For the consumer, one such drawback is that of sizing. Different people have different physical sizes and furthermore different manufacturers produce clothing with different physical dimensions, even when a particular article is said by a manufacturer to be of a certain size. For example, a women's dress size 12 from one manufacturer may be physically completely different to a size 12 of a similar article from a different manufacturer. Accordingly, for a given person, an item of clothing from one manufacturer may fit whereas an article of the same size from a different manufacturer may not fit owing to variations in how clothing is sized. Such variations can be frustrating for a consumer, and can lead to significant costs for a retailer in the form of returns for example.

SUMMARY

According to an example, there is provided a method for generating body measurement data of a user, the method comprising generating image data representing at least one image of the user, the or each image including a calibration portion with at least one known dimension, generating a user representation including multiple user body measurements using the image data. The user representation can be mapped to a set of clothing items of preferred fit. A preferred fit can represent a set of preferred or selected measurements for an article of clothing for the user. The set of measurements can comprise a set of ranges of measurements. A preferred fit can be determined using a pre-selected article of clothing to be worn by the user. In an example, mapping the user representation to a set of clothing items of preferred fit can include determining a set of articles of clothing that are defined by one or more measurements that fall within a predetermined, preselected or preferred range relating to a corresponding measurement. The calibration portion can be an article of clothing to be worn by the user, the article including one or more regions of known dimensions, the or each region configured so as to be identifiable by an imaging apparatus operable to generate the image data. The calibration portion can be an object of known dimensions within the field of view of an imaging apparatus operable to generate the image data. A preferred fit can be determined using a user representation in which the user is wearing a preferred item of clothing, whereby to generate a user representation including multiple user body measurements representing the preferred fit for the user. The method can further include using one or more of the known dimensions of the calibration portion to scale the image of the user, whereby to determine the multiple body measurements. The image data can represent multiple images of the user in differing positions. The image data can be used to generate a composite three-dimensional representation of the user. A user aid to assist the user in positioning themselves for image capture for image data generation can be provided. The user aid can be a representation on a display of the imaging apparatus indicating the position of the user with respect to an optimal user position within the field of view of the imaging apparatus. The user representation can be mapped to a set of clothing items of preferred fit by comparing the multiple user body measurements to multiple measurements defining the sizing of an article of clothing. The set of clothing items can be generated by selecting articles of clothing having a predetermined number of the multiple measurements defining the sizing of the article of clothing within a predefined range of the corresponding user body measurements. Articles of clothing with at least one key measurement defining the sizing of the article of clothing within a predefined range of the corresponding user body measurement can be selected. The user representation can be a set of data representing body measurements of the user and data representing the preferred fit of an article of clothing for the user. The or each image of the user can be processed in order to generate a featureless or generic image representing the user. Image data representing a background image of a location at which the image data representing at least one image of the user is generated can be generated, and the method can further include subtracting the background image of the location from the or each image of the user. The weight of the user can be estimated by using the multiple body measurements to determine user volume and using a measure for body density to calculate the weight. The user representation can be used to produce a bespoke article of clothing for the user. For example, a manufacturer can use the user representation to produce an article of clothing to the user specifications and body measurements, including taking into account a preferred fit for the user. The method can further include determining a colour portion in the at least one image of the user and using the colour portion to define a preferred colour for the user. The method can further include determining a colour portion in the at least one image of the user and using the colour portion to select an article of clothing for the user. According to an example, there is provided a system for selecting a set of articles of clothing for a user, the system including an imaging apparatus to generate image data representing at least one image of the user, a calibration portion of known dimensions, a processor configured to generate a user representation including multiple user body measurements using the image data. The processor can be further operable to map the user representation to a set of articles of clothing of preferred fit for the user. The processor can scale the or each image of the user using the known dimensions to generate scaled image data, and use the scaled image data to determine the multiple user body measurements. The system can include a storage apparatus to store data representing multiple articles of clothing with multiple measurements defining the sizing of respective ones of the articles of clothing. The system can compare the multiple user body measurements with the multiple measurements defining the sizing of the articles of clothing to determine the set of articles of clothing that match the users preferred fit. The processor can compare the multiple user body measurements with the multiple measurements defining the sizing of the articles of clothing to determine the set of articles of clothing for which the multiple measurements are within a predefined tolerance of the multiple user body measurements.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments will now be described, by way of example only, with reference to the accompanying drawings, in which: Figure 1 is a schematic representation of a system according to an example;

Figure 2a is a further schematic representation of a system according to an example;

Figure 2b is a schematic representation of a user from two different viewpoints according to an example.

Figure 3 is a schematic representation of a device according to an example;

Figure 4 is a schematic illustration of the positioning of a device including an imaging apparatus relative to a user according to an example;

Figure 5 is a schematic illustration of an image of a user wearing underwear that has a waistband of known size and some example measurements that can be made by the device according to an example.

Figure 6 is a schematic illustration of an image of a user including an article forming a calibration portion with known dimensions according to an example;

Figure 7 is a schematic illustration of an image of a user including an article forming a calibration portion with known dimensions according to an example;

Figure 8 is a schematic representation of some silhouettes of positions a user might stand in to create a 3D model of themselves according to an example;

Figure 9 is a schematic representation of an image of a user in which various user measurements derived from the image are depicted; Figure 10 is a schematic representation of a user wearing an item of preferred clothing according to an example;

Figure 11 is a schematic representation of a system according to an example; Figure 12 is a schematic representation of two user profiles according to an example; and

Figure 13 is a diagram showing aspects and features that can be determined and leveraged using a user representation according to an example.

DESCRIPTION

Example embodiments are described below in sufficient detail to enable those of ordinary skill in the art to embody and implement the systems and processes herein described. It is important to understand that embodiments can be provided in many alternate forms and should not be construed as limited to the examples set forth herein.

Accordingly, while embodiments can be modified in various ways and take on various alternative forms, specific embodiments thereof are shown in the drawings and described in detail below as examples. There is no intent to limit to the particular forms disclosed. On the contrary, all modifications, equivalents, and alternatives falling within the scope of the appended claims should be included. Elements of the example embodiments are consistently denoted by the same reference numerals throughout the drawings and detailed description where appropriate.

The terminology used herein to describe embodiments is not intended to limit the scope. The articles "a," "an," and "the" are singular in that they have a single referent, however the use of the singular form in the present document should not preclude the presence of more than one referent. In other words, elements referred to in the singular can number one or more, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises," "comprising," "includes," and/or "including," when used herein, specify the presence of stated features, items, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, items, steps, operations, elements, components, and/or groups thereof.

Unless otherwise defined, all terms (including technical and scientific terms) used herein are to be interpreted as is customary in the art. It will be further understood that terms in common usage should also be interpreted as is customary in the relevant art and not in an idealized or overly formal sense unless expressly so defined herein. The space race for fashion online is in sizing individuals so they can buy clothes online with the confidence that they will fit into them perfectly. With over $1 Trillion spent on fashion every year, only 10% of sales in the US happen online due to the sizing issue. Today's fashion industry still relies on small, medium, large etc. and yet brands vary and people vary greatly. Being able to create a 3D model of an individual and their favourite items in their wardrobe and then combing that with the data used to create garments would allow users to find items online and in stores that would fit them perfectly. At present users might stick to a specific brand or return items that do not fit or simply not wear those items again creating a great deal of waste in the industry. It is often said that we wear 20% of our wardrobe 80% of the time and this is often due to the sizing issue. In store users have to search the store for items that are within their size range and then to feel confident they need to try these items on and even then when they get the item home they may not feel it fits them properly or as well as their favourite items. Ordering online is even more difficult and creates the worry that the item might not fit and the need to return the item wastes time and resources for the industry and the individual.

According to an example, a means of creating a three dimensional model (3D model) of a user from their body measurements is provided, in which an imagine apparatus is placed on a platform at a distance that ensures the users whole body is within the devices view frame. The user stands in the frame wearing or sticking something on their body or places an object or printed media within the picture frame that is of a known size and the device takes one or a series of pictures of the users body in different positions or from different viewpoints. The device then creates the 3D model of the user by creating a scale based on the known size of an item on their body, the object or the printed media.

There are many instances whereby users might want to know whether an item of clothing will fit them whether online or in a store without the need to try the item on. There are described herein methods by which a user can use a device to create a 3D model of their own body, a personalised mannequin. The device referred to herein could be a smart phone, digital camera, tablet, a computer or any other device that contains a camera. In particular, in the first aspect, there is provided a method of creating a 3D model of a user's body, the method comprising: the positioning of the device on a platform, that might for instance be an ironing board, that is at or around the mid height of the user, the platform being at a distance from a background, that might for instance be a wall or door, and the camera positioned so it faces the wall and that the users entire body is within the frame of the picture. A calibration portion or item of clothing of known dimensions is provided, which can be worn by the user, or otherwise positioned on or near the user. In the case of an article of clothing that acts as the calibration portion, a piece can have a known size, be known to the device or database or inputted by the user. The piece can be constant all the way around the user's body, for example the waistband of a pair of underwear. Alternatively for example, the user can stick one or more stickers of known size on different parts of their body. For example, stickers can be placed on the user's chest, side and their back. The device then uses the known size to scale the user's entire body and create a 3D model.

In a second aspect there is provided a method of creating a 3D model of a user's body, the method comprising: the positioning of the device on a platform, that might for instance be an ironing board, that is at or around the mid height of the user, the platform being at a distance from a background, that might for instance be a wall or door, and the camera positioned so it faces the wall and that the users entire body is within the frame of the picture, the user then places an object of known size to a database or inputted by the user, for instance a drinks container, or the user fixes a printed out shape of known size on or near the wall within the frame or otherwise at another position within the frame (which can be the field of view of the imaging apparatus of the device). The device then uses the known size to scale the user's entire body and create a 3D model.

Figure 1 is a schematic representation of a system according to an example. The system 100 enables selection of a set of articles of clothing for a user 101. An imaging apparatus 103 is used to generate image data 105 representing at least one image 107 of the user 101. In an example, the imaging apparatus 103 can be a camera module of a mobile device 117 such as a mobile telephone, or the camera module of a tablet or laptop computing device for example. Other alternatives are possible, as will be appreciated, such as a webcam for use with a personal computer or a camera module of a gaming device (portable or otherwise) and so on. A calibration portion 109 of known dimensions is provided. In an example, the portion 109 can be an article of clothing to be worn by the user 101. Alternatively, part of an article of clothing worn by the user 101 can be the portion 109. For example, a waistband 111, such as a waistband on underwear, can be the portion 109. In such a case, the waistband 111 can include at least one known dimension, such as the width D of the waistband, and can extend around the waist of the user 101 so that the portion will be visible no matter which direction the user 101 is facing. The imaging apparatus 103 has a field of view 115.

Figure 2a is a schematic representation of a system according to an example. Elements in common with the system of figure 1 are indicated with the same reference numerals as those used in figure 1. In the example of a system 200 of figure 2, the calibration portion 109 can be an article that is placed in the field of view of the imaging apparatus 103. In this connection, the portion 109 can be an object 113 with at least one known dimension D that can be used to provide scale for an image that is captured using the imaging apparatus 103. Further alternatively, the calibration portion 109 can include one or more stickers or other suitable adhesive item or items that can attach to the user or an item of clothing whereby to be within a field of view of the imaging apparatus 103 and which comprise at least one known dimension to enable image scaling. As noted, imaging apparatus 103 can be part of a device 117. In an example, device 117 includes a processor 119 configured to generate a user representation including multiple user body measurements using image data captured by the imaging apparatus 103. In an alternative example, generation of a user representation including multiple user body measurements using image data captured by the imaging apparatus 103 can be performed remotely from the imaging apparatus 103 and/or device 117. For example, a radio frequency wireless communications module 121 can be used to transmit the image data to remote location 123 where it can be processed to generate the user representation. A network 125, such as the internet for example, can be used to transmit the image data to a remote location.

The processor 119 is further operable to map the user representation to a set of articles of clothing of preferred fit for the user. More particularly, a manufacturer or retailer can host or otherwise provide access to a repository such as a database 300 of articles of clothing on a storage apparatus 399 accessible to the device 117, such as via the module 121 for example. The database 300 can store data representing multiple articles of clothing 301 with multiple measurements 303 defining the sizing of respective ones of the articles of clothing. For example, an entry 305 in database 300 for a shirt can provide multiple measurements 307 associated with the shirt. The measurements can be compared with the user representation to provide a measure representing a potential match for the article in question for the user 101. The measure representing a potential match can be a metric representing the degree to which the multiple measurements associated with the article of clothing in question match with the measurements of the user representation. For example, if measurement for a part of an article of clothing falls within a predefined tolerance associated with a user measurement it can be said to match. The tolerance can be a range for a user measurement that provides an acceptable degree deviation from the measurement itself that the user is prepared to accept.

A predefined number of matching measurements for the various measurements may be required for the article to be selected for the user. In an example, one or more key measurements may be required to match before the article is selected. For example, if a user measurement for chest size is 85cm with a tolerance of +/- 2cm and the corresponding measurement of an article of clothing is 87cm, this can provide a match. However, if a user measurement for waist size is defined as a key measurement which must be matched for the article in question to be selected for the user, and the user measurement for waist size is 75cm +/- 2cm and the corresponding measurement of an article of clothing is 70cm, this would not provide a match. Accordingly, as this is a key measurement, the article would not be selected for the user, even if all other measurements matched.

According to an example, a preferred fit for a user can be used to select an article of clothing. A preferred fit can be determined using an article of clothing that represents a user preference for clothing fit. For example, a preferred article of clothing can be worn by the user 101 and the user representation can be generated on the basis of the user wearing this article of clothing whereby to provide a set of measurements for the user that represents the preferred fit. For example, user 101 can wear a shirt, either that they already own or that is provided for them. The user representation is generated with the user wearing the shirt to provide a set of measurement that represents the preferred fit of a shirt for that user on the basis of the shirt that was worn when the user representation was generated. As described above, an article of clothing can be selected from a repository using the preferred fit user representation. That is, the user representation can be mapped to a set of clothing items of preferred fit by comparing the multiple user body measurements to multiple measurements defining the sizing of an article of clothing. An article of clothing with a preferred fit for a user is therefore an article of clothing that has one or more measurements that are within a range providing a desired fit. This could mean that a preferred fit for an article or clothing is tighter or looser on the user in certain regions than may otherwise be the case if the user's body measurements were used in the absence of any data representing a preferred fit for garments. For example, a user may have a certain waist size that would provoke selection of trousers with a certain waist size that the user did not favour because they were either too tight or too loose. Accordingly, data representing a preferred fit tunes the user measurements to enable selection of articles of clothing that fit with a user's preference for, for example, comfort and/or look. Data relating to or representing the colour/colour palette and any patterns of an article of clothing worn by a user can be generated in order to determine preferred colours for example. For example, in the case of a t-shirt for example, the system as described above can process the image data to determine colours or a range of colours in an image captured of a user. In an example, colour data can be used when showing results for suggested designs to a user based on cut, colour and design for example.

According to an example, the calibration portion 109 is used to scale the image 107. Processor 119 can be used to scale the image 107 of the user 101 using the known dimension(s) of the portion 109, whereby to generate scaled image data 150. The scaled image data 150 can be used to determine the multiple user body measurements. More specifically, and with reference to figure 1, the dimension D is known. Accordingly, the dimensions of the user, such as various user measurements representing the user waist size, chest size and so on can be determined from a captured image of the user 107 including the portion 109 by appropriate scaling as will be appreciated by those skilled in the art.

A table representing measurement data for a garment, in this case a pair of shorts, is provided below. For various sizings of the garment, XS (extra small), S (small), M (medium) and so on, the measurements of various aspects of the article are provided. The measurements represent the manufacturer or retailer measurements for the garment, to which user measurements can be mapped and a match determined.

XS S M L XL

Size / Measurement (cm)

1) waist (relax) 37 39.5 42 44.5 47 2) waistband height 6 6 6 6 6

3) hip (down from waistband) 52 54.5 57 59.5 62

front rise (not include

4) waistband) 26 27 28 29 30

back rise (not including

5) waistband) 30 31 32 33 34

6) thigh width (at crotch ) 28.6 29.8 31 32.2 33.4

7) bottom 24.8 25.4 26 26.6 27.2

8) inseam 22 24.5 27 29.5 32

front pocket (length x

9) height) 18x2 18x2 18x2 18x2 18x2

back pocket (length X

10) height) 15x2 15x2 15x2 15x2 15x2

For example, if a user measurement relating to inseam is determined to be 26.5cm with a tolerance of +/-0.5cm, this is mapped to article size M, which indicates that the inseam measurement is 27cm. Various other user measurements can be compared and matched in this way as desired to determine an article to select for a user. An article of preferred fit can be selected in the same way for user measurements representing a preferred fit for the user.

According to an example, a generated image of a user can be combined with other images of the user in order to provide a composite 3D image 160 of the user from which various user measurements can be derived. Measurements can also be determined from individual images. For example, front, rear and side-on images of a user can be used to determined measurements of a user.

Figure 2b is a schematic representation of a user from two different viewpoints according to an example. The viewpoints represent two scaled images of the user that have been captured and processed using a calibration portion so that user measurements can be determined. In the front view on the left side of the figure, the user's width at the waist from the front is indicated by measurement A. In the side view on the right side of the figure, the user's width at the waist from the side is indicated by measurement B. Considering that, typically, a user's waist can be approximated to be a ellipse, the waist measurement of the user can be determined by:

where a= A/2 and b=B/2. Other formulas to calculate the area and/or perimeter of an ellipse can be used as will be appreciated. In the limit as A=B the waist measurement is simplified to that of the circumference of circle (waist = 2m ). Measurements for other areas of the body can also be calculated in the same way, for example, thigh and neck measurements and so on. In the case of a female bust measurement for example, the approximation can include an additional constant that can be user set to accommodate a preference for fit at that portion of the user's body. Similar constants can be set by a user for any other measurement, and can be used to account for variations in shape from the approximation of the ideal noted above.

The individual images and the composite image can be featureless images. That is, the image captured by device 117 can be used to determine the user measurements for the user, and the measurements can be used to scale a generic image of a person, such as a computer generated image of a person for example. The captured image 107 and the composite image 160 can be rendered as featureless or generic user images in order to provide security for the user. For example, if an image of the user is captured in which the user is wearing underwear, a featureless or generic image can be generated using the user measurements so that the captured image does not need to be displayed. Furthermore, the captured image 107 can be deleted from the device once the user measurements are determined, and only the user measurements and/or the generic image of the user need be stored for future reference for the purposes of selecting articles of clothing.

A composite image can be generated in any number of ways that will be appreciated by the skilled person. For example, feature matching in captured images can be used to stitch images together, and the resultant stitched images can be 'wrapped' or mapped onto a torso/body for example, which can be appropriately scaled using the calibration portion sized using the user measurements, to provide the 3D representation of the user. In an example, user measurements derived from the multiple images are used to appropriately size/scale a generic 3D featureless representation of a person. That is, images of a user are used to determine user measurements, and the user measurements are then used to scale a generic representation of a person. The generic representation can be scaled in various ways in order to provide a suitably accurate representation of the body shape of the user 101, which can then be displayed to the user using a display of the device 117 (not shown), it will be appreciated that for the purposes of article matching, only the user measurements need be provided, and that a 3D representation of other image of the user is not required, but may be used for other purposes such as body shape tracking and other health and fitness related purposes. According to an example, device 117 can include a user aid to assist the user 101 in positioning themselves for image capture for image data generation. For example, the user aid can be a representation on a display of the device 117 indicating the position of the user with respect to an optimal user position within the field of view 115 of the imaging apparatus. In this connection, the user aid can be a generic image of a person that can be positioned in the display at a location that represents the user's position in the field of view. An optimum position can be indicate, such as by an outline of a person on the display, and the displacement of the generic image can be used to determine an amount by which and a direction in which the user should move in order to position themselves in the optimum position for image capture.

Figure 3 is a schematic representation of a device according to an example. As noted above, device 300 can be any suitable device such as a smart phone or other mobile device. Device 300 could be a gaming console linked to a display device such as a television for example, in which images captured using an imaging apparatus of the gaming device are displayed on the display device. An optimum position for user can be indicated using an outline image 301 displayed using a display 303 of the device 300. The current, such as the instantaneous, position of the user can be displayed as an overlaid graphical element in the form of a generic person 305, which can be a featureless or stylised representation of a person for example.

An indication of the displacement, d, of the user's current position from the optimum position with the field of view of an imaging apparatus of the device 300 can be displayed or otherwise provided by the device 300. For example, an arrow can be displayed to indicate that the user should move position in order to position themselves within the optimum position. Alternatively, or in addition, the image 301 can be displayed with a colour such as red to indicate that the user is not within the optimum position. The colour can change to green for example when the use is appropriately repositioned to be substantially at the optimum position. Figure 4 is a schematic illustration of the positioning of a device including an imaging apparatus relative to a user according to an example. User 400 is positioned against a backdrop 401 such as a wall for example. A device 403 is positioned on a support 404 that can be any suitable support that enables the user 400 to be within the field of view 405 of the device 403. An image of the user captured by the device can be processed within the device 403 or remotely. For example, image data 407 generated by the device 403 can be provided to a remote device 409, such a cloud based computing service for example, that can process the image data 407 in order to determine user measurements for example. The user measurements 411 derived from the image data can be provided from the device 409 to device 403. Data between the devices can be transmitted over the internet for example.

Figure 5 is a schematic illustration of an image of a user wearing underwear that has a waistband of known size and some example measurements that can be made by the device according to an example. Image 500 is an image a user 501 in which the user is wearing an item of clothing 502 that includes a calibration portion 505 of known dimensions. As shown in the expanded portion 503 of the image 500, the calibration portion 505 takes the form of the waistband of a pair of underwear that has a known thickness D. The value of D can be used to scale the image 500 in order to determine user measurements.

Figure 6 is a schematic illustration of an image of a user including an article forming a calibration portion with known dimensions according to an example. Similarly to figure 5, an image 600 of a user 501 is provided. An article 602 is provided in the image that has at least one known dimension 603. In the example of figure 6, the article representing the calibration portion is a tin can with known height.

Figure 7 is a schematic illustration of an image of a user including an article forming a calibration portion with known dimensions according to an example. Similarly to figure 6, an image 700 of a user 501 is provided. An article 702 is provided in the image that has at least one known dimension 703. In the example of figure 7, the article representing the calibration portion is a sticker 705 or similar disposable item that can be applied to the background 707 or can be applied to the user 501, such as by being stuck to an item of clothing for example. Figure 8 is a schematic representation of some silhouettes of positions a user might stand in to create a 3D model of themselves according to an example. In the example of figure 8, a user 801 is shown from three different sides - the front 802a, back 802b and side 802c.

Figure 9 is a schematic representation of an image of a user in which various user measurements derived from the image are depicted. An image 900 of a user 901 wearing an article of clothing including a calibration portion 609 with at least one known dimension is provided. Various user measurements 902-908 can be derived from the image 900 of the user 901 and used to provide a user representation that can be sued to select suitable articles of clothing for the user 901 by matching the measurements to known measurements of articles of clothing such as those described above.

Figure 10 is a schematic representation of a user wearing an item of preferred clothing according to an example. In the example of figure 10, an image 1000 of a user 1001 is provided as captured by a device as described above for example. In the image, the user 1001 is wearing an item of clothing 1003 that is of a preferred fit for the user. Accordingly, multiple user measurements of the user 1001 in the item of clothing 1003 can be derived as described above, with the aid of a calibration portion 1005, various types of which are depicted in figure 10 (only one of which may be present in an image proper). The user measurements can be used to select an article of clothing of preferred fit for the user. That is, the user measurements can be used to generate a user representation which can be mapped or otherwise compared with measurement of a multiple clothing items to select a set of clothing items of preferred fit for the user. So, for example, image data is generated using an imaging apparatus of a device as described above, and the image data represents at least one image of the user, which includes a calibration portion with at least one known dimension. A user representation including multiple user body measurements is generated using the image data. For example, the user representation can be a collection of user measurements and can optionally include a user identification that is linked with the measurements. The user identification can optionally include data such as data relating to the age, sex, location and so on of the user that can be used to track the user's measurements over time. The user representation is mapped to a set of clothing items of preferred fit. For example, the user representation can be compared against the measurements of multiple articles of clothing to determine a set of matching articles of clothing that can form the set of clothing items of preferred fit.

In an example, a device such as device 117 for example can receive user input to specify that the user is wearing an item of clothing that represents a preferred fit for example. That is, an indication can be provided to the device that can be used to indicate that the user measurements represent measurements of the user in a preferred item of clothing, and thus the measurements represent the user's preferred fit for such an item. In an example, a tolerance associated with user measurements can be predefined, such as being input by a user, or can be determined using an item of preferred fit for the user. For example, initially, multiple user measurements for a user can be determined using an image of the user in which a calibration portion is provided. Another image of the user wearing an item of preferred clothing can then be captured from which user measurements are derived. The measurements determined from the image in which the user is not wearing the item of preferred clothing and the image in which the user is wearing that item can then be compared in order to generate tolerance values for the user measurements. For example, a user measurement for a user's chest size without the item of preferred fit on may be determined to be 85cm. With the item of preferred fit on, the chest size may be determined to be 90cm. Accordingly a tolerance of +5cm can be associated with the measurement representing the user's chest size, thereby indicating that an article of clothing with a chest measurement that is up to 5cm larger than the chest size as measure is acceptable. In an example, articles of clothing can be rejected (or not selected) if they do not have a measurement that corresponds to the user measurement plus the tolerance value. In the example above, clothing can be rejected if the chest measurement is not at least 90cm for example. In a further example, a user measurement can be set as that which includes the tolerance as noted above, and which includes a further tolerance value. For example, as noted above, if a chest measurement is 85cm, and a preferred chest fit is determined to be 90cm, 90cm can be set as the user measurement for chest size, and a further tolerance value associated with this can be set, such as +/- 2cm for example. Therefore, articles of clothing that have a chest measurement within a range from 88-92cm can be selected as they fall within the preferred range for this measurement. According to an example, a user representation can be stored on a user device for use by the user. For example, a user equipment device, such as a mobile telephone for example, can store a user representation that can be manually provided for the device by the user, such as by being transferred from another device, or pushed to the device from a remote storage location, such as a cloud based service over the internet for example. For example, a dedicated application ("app") can be provided that uses the user representation to modify the proportions of a generic mannequin so that the users body measurements are mirrored by the mannequin, and can be viewed on a display of the device for example. Accordingly, when a user enters or is in the vicinity of a retail store, the user representation can be used to provide the user with a image showing the mannequin in an item of clothing. For example, the retail store can use a beacon that scans for nearby users that are enabled to receive marketing information from the store. A user device in range and opted in can then grant permission for the store to use the user representation, or a portion thereof, which can be used by the store to select an article of clothing and provide data to the device showing that article on the mannequin for example. The selected article can be selected using additional metrics such as previous buying history of the user, thereby taking into account user preference for articles of clothing, potential budget based on previous spending habits and so on. In an alternative example, a store beacon can periodically transmit an invitation to nearby devices which can respond if a user gives permission to do so. For example, a store can transmit data representing an invitation, such as an invitation for a user to provide their user representation. The user device can receive the data and alert the user who may then allow the store to use their representation. In the event that the user sanctions the use, the store can use the representation to provide data representing an exemplary article of clothing that would fit the user and which can be displayed to the user using the device. Such a process can be used to tempt a user into a store to browse and so on, and allow a manufacturer or retailer to promote certain items or ranges of clothing. In an example, the user may also sanction (either on an ad hoc basis or globally) use of other related data, such as an internet browsing history for example from which a retailer can derive buying habit and preferences for example.

Using the user representation, a store can provide a location to enable a user to determine the location of suitable articles of clothing within the store, such as items of preferred fit for example. As is typical, the user device can be used to indicate the location compared to the current location of the device for example. In an alternative example, a location within the store can be used to provide a user with directions to a specific location or locations with the store. For example, a kiosk or other such device within a store can be provided that can communicate with a user device, such as wireless using near field communication for example. The location can read user representation data and use the data to select articles of clothing of a preferred fit for the user for example. The location can then display the items to the user using a display at the location, on the user device or both, and display a location of one or more such selected items at the location, on the user device or both.

In an example, data tags can be provided on garments, such as in or on price labels of garments or on a shelf end display for example. The data tags can be radio frequency identification tags for example that comprise data relating to an article of clothing such as the measurements of various sizing of the clothing as explained above. A user equipped with a suitable device can read data from the tag(s) and use the data in combination with a user representation to determine the suitability of the article of clothing in question.

Accordingly, a consumers profile or user representation can be used in combination with a near field or other suitable wireless communication technology to help them quickly find items in store that would fit them. Changing rooms could also include a screen whereby users no longer need to try on the garment but could see a representation of themselves wearing the garment or their devices. In an alternative, as mentioned above, an article of clothing can be modelled using a mannequin or customised avatar on the device taking into account a user representation.

Typically, retailers in the fashion and more generally the clothing industry are often in- between the manufacturer and the consumer and this is often because retail is predominantly a bricks and mortar business model. Accordingly, an application as noted above can connect user with a manufacturer and facilitate a more efficient business model direct from the manufacturer to the consumer. The manufacturer for a garment could provide dimensions of the garment and that garment could be matched with the consumer without the consumer even needing to search for the item for example. Alternatively, the consumer could provide their up to date measurements to a manufacturer, which could be used to manufacture orders of garments placed by the consumer. In this way the consumer would receive garments on a precise made to measure basis. For example, a user representation can be provided to a manufacturer, and the associated user body measurements can be used to produce a bespoke article of clothing. The user can also provide an indication of selected or preferred materials to be used for the article of clothing to be produced. For example, in the case of a suit to be made to the user's exact measurements, the user can select materials from which the suit is to be made by viewing samples of such materials using the user device which can receive data from manufacturer representing, for example, imagery and/or other associated data for materials that the manufacturer can use for bespoke articles.

Figure 11 is a schematic representation of a system according to an example. More particularly, in the system of figure 11, a user device 1100, such as that described above is used by a user 1101 to provide a user representation 1103 (derived for the user as described above) to a manufacturer 1105. For example, data 1106 representing the user representation 1103 can be transmitted to a server 1107 of or for the manufacturer. The data 1106 can include a user order 1109 specifying the nature of the or each article of clothing that the user 1101 is desirous of. In response to the order 1109, the manufacturer can transmit data 1111 to the user device 1100 representing materials that can be used or which have been selected by the manufacturer for the order 1109. For example, the data can provide details of materials to be used, or a set of material options that can be viewed by the user 1101, such as on a display of the device 1100, to select a material or materials to be sued for the order 1109. Such a material selection can be communicated back to the manufacturer in a response 1113. The manufacturer can optionally acknowledge the message 1113 and production can begin. The user can periodically be sent updates from the manufacturer about the progress of the order, or can retrieve update data 1115 representing the current status of the order 1109 from the server 1107, which can be periodically updated by the manufacturer, or held updated on a real time basis as the article in question progresses through the manufacturing process. According to an example, a user can establish a profile or user representation and their measurements could remain secure but third parties, for instance, family and friends could purchase items for the user confident the garments would fit them. For example, a user can specify that certain aspects or all of a user representation can be shared with selected individuals or devices. Those individuals or devices can then use the representation. A user representation can be used to determine a user's weight. For example, user body measurements can be used to estimate the volume of a user from which the weight of the user can be derived using a measure for the average density of the human body. For example, on average, the human body weighs l .Olg per cubic centimetre. Accordingly, a measure for volume of a user yields an estimated user weight.

Figure 12 is a schematic representation of two user profiles according to an example. More particularly, figure 12 depicts a user from the side (left hand side of figure 12), and the front (right hand side of figure 12). Multiple measurement lines are depicted in the figure for each user profile. Each horizontal line represents the width of the user at the point in question. With reference to the above, the width measurements at the different points depicted can be used to determine the circumference of an ellipse at each point that can provide a measure of the user's circumference at that point or of the body part in question (such as arms and legs for example). The circumference measurements can then be used to determine the user volume. For example, successive circumference measurements multiplied by the vertical distance there between yield a measure of volume for a user segment. Multiple such volume measures can be determined and used to provide an overall volume for the user. In the limit, for example at the pixel level of the images of the user in question, each circumference measurement can be used to determine the cross-sectional area, A, of the user at that point according to:

A = nab

As above, a= A/2 and b=B/2, where A and B represent the lengths of the major and minor axes of an ellipse. The determined areas can be summed to determine user volume, for example using height measurements (pixel values) at various points of the users body so as to enable volume portions to be determined which can be added together to provide an overall volume measure of the user. For example, with reference to figure 12, multiple vertical (or off -vertical) lines are depicted that represent heights at various portions of the user's body, and which can be used for volume calculations.

In an example, the measurement lines, such as 1201, 1203 for example, can be one pixel in height in order to maximise accuracy. The measurement lines can thus be packed closer together, such as with only a pixel between lines for example to provide multiple measurements for the user. In the example of figure 12, larger spacing is shown for the sake of clarity. The vertical or off vertical lines, such as 1205, 1207 for example, can be used to determine user height along their respective lengths, which can be used as noted above to determine user volume, and thus user weight. Thus, in relation to figure 12 and user weight, the horizontal lines can, for example, be 1 pixel apart, and hence provide a more accurate measure across the height of the user than the lines shown in the figure. The more horizontal lines the more accurate the volume would be and ultimately the entire model. That is, the measurement lines shown in figure 12 are not intended to be limiting. In an example, an image of user can be considered at the pixel level. For example, the lines depicted in figure 12 can be one pixel in height or width. Accordingly, using each measurement line to determine the area of an ellipse as noted above, each ellipse can represent a slice or layer for the user. The layers can be stacked to form a 3d mannequin of the user. The more layers that are used, the more precise the model will be.

Figure 13 is a diagram showing aspects and features that can be determined and leveraged using a user representation according to an example. Data 1301 representing a user representation that has been generated for a user as described above is provided. As noted, body measurements forming part of the user representation can be used to scale a 3D representation of the user. As noted above, the 3D representation can be a generic (featureless) mannequin for example, which is proportioned in line with the user's body measurements in order to provide an accurate (albeit generic) representation of the user's shape and proportions. Alternatively, the images used to determine the user's body measurements can be used to provide a scaled 3D representation of the user, as noted above. More particularly, multiple two-dimensional images can be stitched together, such as by determining image features using known techniques and using the features to align the images by matching for example.

A 3D body image, and more generally user representation data can be used in various ways according to an example. These can broadly be compartmentalised into fashion related uses 1303, health related uses 1305 and fitness related uses 1307. Sector 1303 in figure 13 depicts some aspects according to an example. For example, a 3D body image 1309 of a user can be create a 3D mannequin 1311 as described above. The mannequin 1311 can be used to determine a user's favourite clothes 1313 by allowing a user to see how articles clothing look whilst they are being worn by the mannequin. For example, image data relating to articles of clothing can be used to add clothing to the mannequin such that it appears as if the user is wearing the article(s) question. Favourite items of clothing can be used to perform manufacturer searches for articles with the same or similar dimensions for example. Data representing the 3D mannequin can be used by 3^rd parties to provide the user with examples of clothing being worn by the user. For example, an API 1315 can provided to a 3^rd party to enable them to use the mannequin and/or user representation.

In a health related sector 1305, a body image or data for a user representation can be used to determine a user's weight, as described above, and thus used effective as a form of digital weighing scales 1317. Weight information and more generally a user representation can be used to profile a user's fat composition or percentage. Fat storage areas can be determined from an image to develop a risk assessment 1319. For example, areas of a user's body that deviate to a predetermined degree from a norm or accepted range can be used to determine the risks associated with particular conditions. For example, if a user image or representation indicates that a user has an above average waist size for their height and so on, this can be used to present an assessment relating to certain conditions that can be adversely affected by such a factor. Similarly, an image can also be used to determine the extent of water retention 1321 of a user. In the same sector, a metabolism assessment 1323 can also be provided. For example, a user's weight can be determined, and over time this can be compared to previous weights, which in combination with a calorie or other such record for the user can be used to determine a measure for the user's metabolism. More generally, in the fitness related sector 1307. a user's body shape can be monitored over time using a body shape monitor 1325. For example, the size and/or weight of a user can be tracked over time. Size can be a measure of volume, or a measure of one or more user body measurements for example. In combination with weight and metabolism data 1327 this can be used to provide diet advice for a user, such as by way of an internet enabled smart phone executing an associated app for example.

Body shape monitor 1325 can be sued to provide fitness advice 1329. For example, over time, if certain areas of a user's physique are changing at a rate that above or below a predefined threshold or norm, fitness advice can be provided to enable the user to tailor exercises to either increase or decrease the rate of change for the area in question, for example, to increase the rate of weight loss at the waist area of a user. The change for the area in question can relate to fat content, size, volume and so on. Data 1331 from 3^rd parties can be provided to augment the fitness advice. For example, data representing a specific exercise to increase or decrease the rate of change in an area of the user's body can be used.

Body image 1309 and/or data 1301 can be provided to 3^rd party services such as social media services or applications 1333 for example. In an example, a user can upload the relevant image or data to a social media portal that can display certain aspects or all of the data, as the user desires or sets, to other users who may or may not be associated with the user in question (e.g. 'friends' or more generally any user of the social media platform in question).

According to an example, once a user representation has been generated for a user using a scaled item in shot, a user profile can be created and saved on the user device and/or a server, such as a server noted above. A particular a user dimension, for instance distance between their eyes (front view) or ears (side view) can be saved for future scaling so that a calibration portion does not need to be provided for future images of the user. That is, a user measurement derived when a calibration portion is present can, in essence, be used as a calibration portion for other images of the user.

A combination of measurements can be used to provide a measurement ID for the user. For example, the measurements for the distance between the user's eyes and their waist measurement can be combined using a predefined function and hashed using a has function to provide an ID for the user that can be used for recognition of the user and/or future scaling purposes. For example, if a trusted third party is provided with or knows the function used to generate the combination or the hashing function it can determine the measurements that can be subsequently be used. Alternatively for example, an image of a user can be used to determine body measurements that can be combined. The combinations, such as a hash, can be compared against a known trusted value to determine if the user is an expected user, which can provide a level of security.

Claims

1. A method for generating body measurement data of a user, the method comprising:

generating image data representing at least one image of the user, the or each image including a calibration portion with at least one known dimension;

generating a user representation including multiple user body measurements using the image data.

2. A method as claimed in claim 1, further including mapping the user representation to a set of clothing items of preferred fit.

3. A method as claimed in claim 2, wherein a preferred fit represents a set of preferred or selected measurements for an article of clothing for the user.

4. A method as claimed in claim 3, wherein the set of measurements comprises a set of ranges of measurements.

5. A method as claimed in any of claims 2 to 4, wherein a preferred fit is determined using a pre-selected article of clothing to be worn by the user.

6. A method as claimed in any preceding claim, wherein the calibration portion is an article of clothing to be worn by the user, the article including one or more regions of known dimensions, the or each region configured so as to be identifiable by an imaging apparatus operable to generate the image data.

7. A method as claimed in any of claims 1 to 5, wherein the calibration portion is an object of known dimensions within the field of view of an imaging apparatus operable to generate the image data.

8. A method as claimed in any preceding claim, wherein a preferred fit is determined using a user representation in which the user is wearing a preferred item of clothing, whereby to generate a user representation including multiple user body measurements representing the preferred fit for the user.

9. A method as claimed in any preceding claim, further including:

using one or more of the known dimensions of the calibration portion to scale the image of the user, whereby to determine the multiple body measurements.

10. A method as claimed in any preceding claim, wherein the image data represents multiple images of the user in differing positions.

11. A method as claimed in claim 10, further including:

using the image data to generate a composite three-dimensional representation of the user.

12. A method as claimed in any preceding claim, further including:

providing a user aid to assist the user in positioning themselves for image capture for image data generation.

13. A method as claimed in claim 12, wherein the user aid is a representation on a display of the imaging apparatus indicating the position of the user with respect to an optimal user position within the field of view of the imaging apparatus.

14. A method as claimed in any preceding claim, wherein the user representation is mapped to a set of clothing items of preferred fit by comparing the multiple user body measurements to multiple measurements defining the sizing of an article of clothing.

15. A method as claimed in claim 14, wherein the set of clothing items is generated by selecting articles of clothing having a predetermined number of the multiple measurements defining the sizing of the article of clothing within a predefined range of the corresponding user body measurements.

16. A method as claimed in claim 15, further including selecting articles of clothing with at least one key measurement defining the sizing of the article of clothing within a predefined range of the corresponding user body measurement.

17. A method as claimed in any preceding claim, wherein the user representation is a set of data representing body measurements of the user and data representing the preferred fit of an article of clothing for the user.

18. A method as claimed in any preceding claim, wherein the or each image of the user is processed in order to generate a featureless or generic image representing the user.

19. A method as claimed in any preceding claim, further including generating image data representing a background image of a location at which the image data representing at least one image of the user is generated, and subtracting the background image of the location from the or each image of the user.

20. A method as claimed in any preceding claim, further including estimating the weight of the user by using the multiple body measurements to determine user volume and using a measure for body density to calculate the weight.

21. A method as claimed in any preceding claim, further including using the user representation to produce a bespoke article of clothing for the user.

22. A method as claimed in any preceding claim, further including determining a colour portion in the at least one image of the user and using the colour portion to define a preferred colour for the user.

23. A method as claimed in any of claims 1 to 21, further including determining a colour portion in the at least one image of the user and using the colour portion to select an article of clothing for the user.

24. A system for selecting a set of articles of clothing for a user, the system including: an imaging apparatus to generate image data representing at least one image of the user; a calibration portion of known dimensions;

a processor configured to generate a user representation including multiple user body measurements using the image data.

25. A system as claimed in claim 24, further operable to map the user representation to a set of articles of clothing of preferred fit for the user.

26. A system as claimed in claim 24 or 25, wherein the processor is further operable to scale the or each image of the user using the known dimensions to generate scaled image data, and to use the scaled image data to determine the multiple user body measurements.

27. A system as claimed in any of claims 24 to 26, further including a storage apparatus to store data representing multiple articles of clothing with multiple measurements defining the sizing of respective ones of the articles of clothing.

28. A system as claimed in claim 27, the system operable to compare the multiple user body measurements with the multiple measurements defining the sizing of the articles of clothing to determine the set of articles of clothing that match the users preferred fit.

29. A system as claimed in claim 28, wherein the processor is operable to compare the multiple user body measurements with the multiple measurements defining the sizing of the articles of clothing to determine the set of articles of clothing for which the multiple measurements are within a predefined tolerance of the multiple user body measurements.

30. A method substantially as hereinbefore described with reference to and as shown in the accompanying drawings.

31. A system substantially as hereinbefore described with reference to and as shown in the accompanying drawings.