Conference PaperPDF Available
Kidsize: Always Get the Right Size!
Alfredo BALLESTER, Ana PIÉROLA, Eduardo PARRILLA, Jordi URIEL, Paola PIQUERAS,
Beatriz NÁCHER, Julio VIVAS, Cristina PÉREZ, Silvia SAN JERÓNIMO, Sandra ALEMANY
Instituto de Biomecánica (IBV), Valencia, Spain
DOI: 10.15221/16.139 http://dx.doi.org/10.15221/16.139
Abstract
This paper describes the two innovations underpinning Kidsize concept and presents the results of
their validation. The first one is a mobile phone app to measure a child in 3D by taking two pictures.
This new method is more accurate and consistent than an untrained person using a measuring tape at
home or in the shop. The second one is an expert system that recommends the size that best fits the
child and assesses the fit of the garment at different body areas. Project results show that it can
provide nearly 90% right size recommendations, thus outperforming existing methods like age- or
height-based size guides, which achieve 40 and 60% right recommendations respectively.
Keywords: 3D scanning, anthropometry, body models, avatar, mannequin, children, childrenswear,
size, apparel, garment, clothing, fit, online, measurement, shape, data-driven, phone, tablet, app.
1. Introduction
Getting the right size without trying the clothes on is a challenge [1] that every parent faces whenever
they intend to buy clothing for their children, either online or at brick-and-mortar stores. Children’s
sizing has the particularity of being designed based on statures but being labelled according to age (i.e.
months and years); even though children of the same age may have different dimensions and body
proportions. Moreover, parents do not always take their children with them to the shops and when they
do, they rarely try the garments on them. The problem becomes even more acute if it is a friend or
relative who wants to make a gift. These facts, together with the differences in the growing pace of
children, create confusion among buyers when they have to pick a size for a child.
In this context, Kidsize project [2 , 3 ] aimed at developing technologies to help childrenswear
manufacturers and retailers to sell more, by clearing up the size-related concerns when buying online
or at the retail stores when parents buy without their children, and to sell better, by cutting down the
expenses derived from product returns both at online and retail stores.
In particular, Kidsize targeted at providing parents (or relatives) with the information required to pick
the right size for their children. In the Kidsize concept, this information included:
a) a size recommendation for wearing the garment straight away (herein “expert’s advice”)
b) a size recommendation for the best fit to allow room for the child to grow (herein “parents’ advice”)
c) how the garment is expected to fit the child at relevant body areas like the shoulder, the chest, the
hips, the sleeve length or the trouser length (herein “fit-by-area”)
In order to provide this information, it is necessary to establish rules/models that explain/simulate the
user-product interaction [4,5,6,7,8], which in our childrenswear context means “wearing the garment
comfortably with an aesthetically nice fit”. If such rules, models and/or simulations are expected to be
somehow reliable, it is necessary that they are based on reliable information/properties from the
person (e.g. body dimensions, shape or preference) and from the garment (e.g. dimensions, style, type
of garment or materials).
Reliable information/properties of the garment can be available to the producers or merchants that
have access to the product (or to its design) along the supply chain. Any producer or brand which has
a narrow control of the conception and/or of the quality control of their products should have access to
it.
In contrast, a regular person does not have access to reliable dimensional information of him/herself or
of his/her children (except for, maybe, body height and weight). The two methods for reliably and
consistently measuring the human body are: being measured by an expert [9,10] or being 3D-scanned.
Unfortunately, nowadays none of these methods are readily available when we buy online at home or
when we go shopping to brick-and-mortar stores.
* alfredo.ballester@ibv.upv.es; +34 610 562 532; http://anthropometry.ibv.org
Proceedings of the 7th International Conference on 3D Body Scanning Technologies, Lugano, Switzerland, 30 Nov.-1 Dec. 2016
139
Within this context, the R&D work conducted under Kidsize project (figure 1) was focused on three
scientific and technical challenges:
1) Developing a reliable body measuring instrument that could be used almost anywhere, at any
time and by anyone (i.e. that does not require any special expertise).
2) Developing algorithms that could provide a right size advice in most of the cases, i.e. 90%.
3) Implementing a prototype solution adapted to the needs of the stakeholders involved in the
purchase of childrenswear, i.e. the buyers, the manufacturers and the retailers.
Figure 1: Kidsize concept incl. the main back-end and front-end components
This paper focuses on the first two objectives. Section 2 of this paper describes the body measuring
instrument and the size recommendation technology. Section 3 describes the validation studies
conducted within the project and their results and Section 4 summarises the conclusions drawn.
2. Description of the technologies
2.1. Kidsize app – a new body measuring instrument
The access to the 3D representation of people’s body would be the best source of shape and
dimensional information about the body. However, there are several barriers that have hindered the
massive spread of 3D scanners as consumer goods or as a typical in-store appliance: the price is high
and devices are too bulky for homes and retail stores; and many of them require expertise to achieve a
quality scan and to locate the anatomical references to get the right measurements.
In Kidsize we developed an approach for estimating the 3D shape of a child’s body just by taking
him/her a couple of pictures using a smartphone application (figure 2). That way, the body measuring
instrument could be available at any place with internet access.
Figure 2: Kidsize app for phones and tablets
Proceedings of the 7th International Conference on 3D Body Scanning Technologies, Lugano, Switzerland, 30 Nov.-1 Dec. 2016
140
Our approach is based on data-driven methods for 3D reconstruction [11,12]. For children that cannot
stand or follow instructions (until 3 years-old) our method uses a Partial Least Square (PLS) regression
to obtain 19 body dimensions from age, gender, height and weight reported by parents. The regression
model was built from a database of 276 children aged 0-3 y.o. measured by an expert using traditional
methods.
For children from 3 years-old, our approach is based on segmenting the body outlines from two images
and then optimising a parameterised body shape model until the outlines of its projections match the
segmented outlines from the images [13]. The children’s body shape model was obtained by applying
Principal Components Analysis (PCA) [14,15] to a database of 761 registered 3D scans of children
aged 3-12 y.o. [16]. The two photographs (front and side) are taken with the child standing, wearing
tight clothing (or swimwear) in front of a clear wall. In order to facilitate this process, the app provides a
guiding silhouette. For segmenting the body figure from the background we used an adaptation of
Grabcut algorithm [17]. The body height of the child reported by the parents and the camera
parameters of the phone/tablet are used to calibrate the images. The 3D reconstruction optimization is
computed on the cloud but the segmentation of the images is made in the phone so that actual children
images are not sent through the Internet. The app was implemented for Android 4.4+ phones/tablets.
2.2. Kidsize algorithms for giving size advice and estimating the garment fit
Kidsize provides size advice (expert’s and parents’) and fit predictions from a set of body
measurements of the child and the properties of the garment (figure 3). In order to model and explain
the child-garment interaction we used Ordinal Logistic Regression (OLR) with stepwise variable
selection and a set of expert rules using an adaptation of the methods proposed by Alemany et al. [1].
Figure 3: Kidsize’s size advice and fitting prediction
The OLR models were trained using experimental data from over 1100 garment evaluations (fit trials).
11 garments of different types (incl. bodysuits, rompers, trousers, t-shirts, shirts, skirts and dresses)
where provided in all their size spans by two childrenswear brands (i.e. Lullaby and Bóboli). 160
children aged 0-12 y.o. and their parents voluntarily participated in the fit trials (figure 4). Prior to the fit
trials the participants were measured using traditional methods (0-3 y.o.) or using Vitus XXL 3D body
scanner (3-12 y.o.). 19 and 36 body dimensions were measured respectively for babies and children.
Figure 4: Images of the fit trials
Proceedings of the 7th International Conference on 3D Body Scanning Technologies, Lugano, Switzerland, 30 Nov.-1 Dec. 2016
141
At each fit trial session, the child tried on at least two garments in at least two sizes. At each fit trial, the
size was evaluated by a fashion expert using a three-level scale (small, correct or big) and by parents
who declared if they would pick the size or not. Moreover, the expert also assessed the fit at the
relevant body areas using a five-level Likert scale ranging from very tight/short to very loose/long. The
relevant body areas defined were different depending on the type of garment (figure 5).
Figure 5: examples of the relevant body areas for a dress and a shirt
A total of 12 different OLR models (equation 1) were defined according to: the type of advice (expert
and parents), the age market segment (babies 0-3 y.o. and children 3-12 y.o.) and the type of garment
(upper body, lower body and full body). Two models (one for babies and one for children) were also
defined for each of the 19 body fit areas (38 fit-by-area models in total). The original measurement sets
were reduced to 9 variables by the stepwise variable selection (figure 6).
Equation 1: General formula and probability formulae for size advice prediction
Figure 6: set of 9 body measurements selected for children and babies’ OLR models
The expert rules that applied to the output probabilities of the models were very simple:
The recommended size was the biggest size with good fit.
If all sizes were big or all sizes were small, the system would not recommend any size.
If the child was between two sizes (there is no size with a clear good fit), the recommended
size was the bigger of the two.
The parents’ advice was always equal or one size bigger than the experts’.
























Proceedings of the 7th International Conference on 3D Body Scanning Technologies, Lugano, Switzerland, 30 Nov.-1 Dec. 2016
142
3. Validation studies and results obtained
3.1. Precision of Kidsize app (repeated measurements)
In order to assess and quantify the precision and consistency of the body measurements obtained
using Kidsize app, we conducted a study using scale figurines. Six synthetic children shapes
representing the boundaries of the body shape space were generated and manufactured in 1:10 scale
using a Solid Laser Sintering (SLS) machine from EOS (figure 7). Each of the figurines was
photographed 10 times in a contrast controlled environment. Its actual height was used as scale up
parameter for the 3D reconstruction.
Figure 7: virtual models, scale figurine set, controlled background and picture taking process
The body measurements obtained from the app were compared to the actual measurements of the
virtual models. To evaluate the error in measurement estimation, the Mean Absolute Difference (MAD)
for repeated measurements was computed (equation 2), where is the number of figurines and
is
the number of repetitions for figurine , in this study (10 repetitions),
 for all .


 



Equation 2: Mean Absolute Difference (MAD) for repeated measurements
Table 1 gathers the results obtained for the measurements used in size advice and compares it to the
results obtained by different studies of precision of high resolution 3D body scanners in adults.
Table 1: Precision (MAD in mm and %) of Kidsize app with scale figurines compared to 3D scanner precision
results from bibliography (Lu & Wang, 2010; Dekker, 2000; Robinette & Daanen, 2006)
Measurement name
Kidsize
MAD
Kidsize
MAD %
3D scanner
MAD*
3D scanner
MAD [18]
3D scanner
MAD [19]
3D scanner
MAD [20]
Knee height
3
1%
3
1
4
Mid neck girth
5
2%
3
Chest girth
8
1%
10
2
6
Back armpits contour
6
2%
8
1
Seat
girth
6
1%
4
2
Cervical height
3
0%
3
4
Waist girth
8
1%
5
3
3
Arm length
6
1%
5
5
5
3.2. Accuracy of Kidsize app in babies (compared to expert measurements)
In order to determine the accuracy of Kidsize App for babies compared to traditional anthropometric
methods, we conducted a study with 30 children aged 0-3 y.o. Each child was measured using both
methods. An anthropometry technician (expert) measured the babies for 19 measurements plus body
mass and body length. A parent (non-expert) measured body mass and length and used them to get
the 19 measurements with the app.
* Results obtained by IBV in non-published results of repeatability tests (4 repetitions) with children using Vitus
XXL 3D body scanner and own developed tools for posture harmonisation and digital body measuring [15].
Proceedings of the 7th International Conference on 3D Body Scanning Technologies, Lugano, Switzerland, 30 Nov.-1 Dec. 2016
143
Firstly, we analysed the differences between parents’ and expert’s input, i.e. body mass and body
length, using a paired t-test of the mean difference. Test results showed that there were no significant
differences (p<0.05) between weight and height measured by parents or experts. The differences of
the averages observed were of -0.2 Kg for body mass and of -0.2 cm for body height. Secondly, we
calculated the Mean Absolute Difference (MAD) to have an estimate of accuracy for the predicted
measurements (equation 3), where is the number of babies.



Equation 3: Mean Absolute Difference (MAD)
Table 2 summarises the results obtained for the measurements that are used in the size advice for
babies and it is compared to bibliographic results of measurements taken by non-experts.
Table 2: Accuracy (MAD in mm and %) of Kidsize app for babies (Kidsize vs. expert) and its comparison to the
accuracy of measurements taken by non-experts (non-expert vs. expert) from bibliography (Yoon & Radwin, 1994)
Measurement name
Kidsize MAD
Kidsize
MAD %
Non
-
expert
MAD [
10
]
Mid neck girth
14
6%
22
Chest girth
12
3%
24
Cervical height
14
2%
35
Waist girth
15
3%
35
Head girth
12
3%
-
Hip girth
18
4%
53
Belly girth
15
3%
24
Wrist girth
8
7%
24
Thigh girth
13
5%
35
3.3. Accuracy of Kidsize app in children (compared to high resolution 3D scanners)
In order to determine the accuracy of Kidsize App for children compared to 3D body scanners, we
conducted a study with 34 children aged 3-12 y.o. Each child was scanned with Vitus XXL 3D scanner
and measured using our data-driven 3D reconstruction app. To evaluate the error in measurement
estimation using the app compared to body scanner, the Mean Absolute Difference (MAD) for
repeated measurements was calculated (equation 3), where is the number of children.
Table 3 summarises the results obtained for the measurements that are used in the size advice for
children and it is compared to bibliographic results of the accuracy of measurements taken by
non-experts.
Table 3: Accuracy (MAD in mm and %) of Kidsize app for children (Kidsize vs. 3D scan) and its comparison to the
accuracy of measurements taken by non-experts (non-expert vs. expert) from bibliography (Yoon & Radwin, 1994)
Measurement
name
Kidsize MAD
Kidsize MAD %
Non
-
expert MAD [
10
]
Knee height
10
3%
41
Mid neck girth
11
4%
22
Chest girth
21
3%
24
Back armpits contour
20
7%
15
Seat girth
12
2%
53
Cervical height
11
1%
35
Waist girth
18
3%
35
Arm length
13
3%
24
3.4. Reliability of the size advices provided by Kidsize
In order to assess the reliability of the size advices provided by Kidsize Solution, we made a test
involving a group of volunteers meeting the target customer profile: parents with one or more children
that sometimes buy childrenswear online. 30 children (23 parents) participated in the test. They were
selected to have a gender- and age-balanced sample (10 children aged 0-2 years old, 10 children
aged 3-7, and 10 children aged 8-12).
Proceedings of the 7th International Conference on 3D Body Scanning Technologies, Lugano, Switzerland, 30 Nov.-1 Dec. 2016
144
The childrenswear brands participating in the project (Bóboli and Sucre d’Orge) provided 19 different
garments (figure 8) in all the available sizes for the testing. The two brands activated the products in
Kidsize system by entering garment properties using the prototype back-office for brands.
Figure 8: sample of garments used in the test (incl. babies’, boys’ and girls’ garments)
The tests took place at two locations: at a Bóboli shop located in a mall (C.C. Bonaire) in Valencia
(figure 9) and at a simulated multi-brand shop at IBV facilities. We prepared a fully working online
demo shop including the Kidsize add-on button in the product selection page so that the participants
could use it on a tablet during the tests (figure 9). Temporarily, Kidsize app was uploaded to Google
Play in order to facilitate its distribution to the testers (figure 9).
Figure 9: Bóboli shop, online demo shop, Android app and iFrame at online shop
During the test, each participant was asked to use the Kidsize app to measure his/her child and to go
shopping four garments at the online demo shop. The four selected garments were randomly assigned
to each participant according to a balanced design of experiments. For each of the garments evaluated,
the parents tested first the size (or sizes) advised by Kidsize (expert’s and parents) and were
encouraged to test other sizes. Then, the parents were requested to pick one size for wearing the
garment straight way and to pick the size that they would buy (usually letting room for the child to
grow); they were allowed to select the same size in both cases. At the end of the test, they were also
requested to answer some questions about their experience using Kidsize.
The reliability of the size advice provided by Kidsize was compared with the available alternatives, i.e.
a size guide based on children stature provided by the brands, and the labelling of garment sizes,
which uses the ages as reference. Results are presented in Table 4.
Table 4. Reliability of size advices provided by Kidsize compared to state of the art solutions
Size Guide (stature) Garment label (age) Kidsize
Expert’s advice 54% 42% 85%
Parent’s advice 59% 48% 88%
Regarding the subjective evaluation, Kidsize solution was positively assessed by parents, who found it
very promising for resolving their concerns when buying without trying on the garments. In general, it
was considered very useful, reliable and easy-to-use. Some of the respondents requested a similar
tool for adults.
4. Discussion and conclusions
Most of the size recommendation systems available in the market use body measurements reported
by the users. However, body measurements, apart from body height and weight, taken by a non-expert
using a measuring tape are not reliable providing an accuracy of 2-5 cm [10]. 3D scanners are much
more reliable instruments for this purpose [18,19,20], but they are not available at most homes or
shops.
Proceedings of the 7th International Conference on 3D Body Scanning Technologies, Lugano, Switzerland, 30 Nov.-1 Dec. 2016
145
According to our results, Kidsize app can provide an accuracy of 1-2 cm and a precision of <1cm. This
makes our new method more accurate than an untrained person using a measuring tape and almost
as consistent as a 3D body scanner. These values seem adequate for performing size
recommendations because they lay within half of the size step for most of the body dimensions.
Moreover, a mobile application can be used to measure a child at home or at any shop.
Most of the returns of childrenswear are reported to be due to wrong sizes. This problem affects both
the online and the brick-and-mortar channels because parents or relatives do not bring the children to
the shops, and when they do they do not try on the garments. According to our results, the size guides
that are available at the points of sale of childrenswear fail to give a right advice in nearly half of the
cases (48-59% right recommendations).
According to our results, Kidsize was consistent with parents’ opinion for 85-88% of the size
recommendations made, clearly outperforming the size guides.
The reliability of the size advices provided by Kidsize showed a similar performance than previous
studies conducted with adult female garments (86-93%) using also OLR for modelling the
user-garment try-on interaction [1]. The performance results obtained by unidimensional size guides
(stature- or age-based) in this study (48-59%) are also similar to those obtained by tri-dimensional size
guides (bust-waist-hip) used in the female study (41-64%).
The differences between the performance of OLR-based methods and the size guides could be
explained by two reasons. Firstly, because the product properties used in the children and female
studies were actually measured or verified, while the size guides published by brands may not
correspond to actual product properties. Secondly, it is possible that more than 2-3 measurements are
required to model the try-on of garments and provide a reliable size advice. The number of variables
used in existing size guides (and in many size recommendation systems available in the market)
ranges from 1 to 3, while the number of body dimensions used in Kidsize ranged from 5 to 9 and in the
female study they ranged from 6 to 20.
Kidsize innovations showed to be very promising but several improvements should be made before
shifting from prototypes to products. Further work will include extending the 3D body reconstruction to
adults and developing size recommendation and fit-by-area algorithms for adult garments. Moreover,
the forthcoming availability of depth sensors in mobile phones and tablets [21,22] will bring an
opportunity to refine and improve data-driven 3D reconstruction technologies.
Acknowledgments
The authors thank the European Commission, the Spanish Ministry of Industry and all the companies,
organisations and volunteers participating in Kidsize [2] and Talla-Me [23] projects. In particular the
authors thank Children’s Fashion Europe (CFE), the French Innovation Network for the Well-being of
Children (NovaCHILD), the Spanish Association of childcare Products (ASEPRI), Star Textil S.A.
(Bóboli), Bravotex (Lullaby), Group Salmon Arc-en-ciel (Sucre d’Orge), Primer Puesto S.L. (Ozongo),
the UK Intelligent Systems Research Institute (ISRI), the University of Valencia (UV) and Jaume I
University of Castellón (UJI).
References
[1] Alemany, S.; Ballester; A., Parrilla; E., Uriel, J.; González, J.; Nácher, B.; González, J.C.; Page, “A.
Exploitation of 3D body databases to improve size selection on the apparel industry”, in Proc. of
4th Int. Conf. on 3D Body Scanning Technologies, Long Beach, CA, USA, November 2013.
[2] "Development of a new extended product-service to overcome size assignment and fitting barriers
for children fashion on-line market addressing customer needs" (KidSize), FP7, European
Commission, FP7-SME-2013-606091, 2014-2016, www.kidsizesolution.com
[3] Kidsize at YouTube, www.youtube.com/channel/UC9Myhl722CSGlSIWKER3H0g (29/09/2016)
[4] HFES 300 Committee, ed. Guidelines for using anthropometric data in product design, Human
Factors and Ergonomics Society, 2004
[5] Loker, S., Ashdown, S. and Schoenfelder, K., "Size-specific analysis of body scan data to improve
apparel fit", Journal of Textile and Apparel, Technology and Management 4.3 (2005): 1-15
Proceedings of the 7th International Conference on 3D Body Scanning Technologies, Lugano, Switzerland, 30 Nov.-1 Dec. 2016
146
[6] Robinette, K. M., "Anthropometry for product design." Handbook of Human Factors and
Ergonomics 4, 2012, pp 330-346.
[7] Robinette, K. M., and Veitch, D., "Sustainable Sizing", Human Factors: The Journal of the Human
Factors and Ergonomics Society, 2016, 0018720816649091
[8] Gill, S., “A review of research and innovation in garment sizing, prototyping and fitting”, Textile
Progress, 47:1, 1-85, 2015, DOI: 10.1080/00405167.2015.1023512
[9] Gordon, C. C., Bradtmiller, B., Churchill, T., Clauser, C. E., McConville, J. T., Tebbetts, I. O., and
Walker, R. A., “1988 Anthropometric Survey of US Army Personnel: Methods and Summary
Statistics”, Natick, MA, US Army Natick Research Development and Engineering Center, 1989.
[10] Yoon, J.C., and Robert G. R. “The accuracy of consumer-made body measurements for women's
mail-order clothing”. Human Factors: The Journal of the Human Factors and Ergonomics Society,
Vol. 3, No. 3, pp. 557-568, 1994.
[11] Parrilla, E.; Ballester, A.; Solves-Camallonga, C.; Nácher, B.; Puigcerver, S.A.; Uriel, J.; Piérola,
A.; González, J.C.; Alemany, S., “Low-cost 3D foot scanner using a mobile app”, Footwear
Science, Vol. 7, Iss. sup1, 2015.
[12] Ballester, A., Parrilla, E., Vivas, J. A., Piérola, A., Uriel, J., Puigcerver, S. A., Piqueras, P.,
Solves-Camallonga, C., Rodríguez, M., González, J. C., and Alemany S, “Low-Cost Data-Driven
3D Reconstruction and its Applications”, In Proc. of 6th Int. Conf. on 3D Body Scanning
Technologies, Lugano, Switzerland, October 2015. doi:10.15221/15.184
[13] Parrilla, E., Ballester, A., Uriel, J., Piérola, A., Pérez, C., Piqueras, P., Nácher, B., Vivas, J. A. and
Alemany, S., “Data-driven 3D reconstruction of human bodies using a mobile phone app” , Int. J.
of the Digital Human, Special Issue on 3D Anthropometric Databases and Their Applications
(submitted 30/06/2016, under review).
[14] Allen, B., Curless, B., and Popović, Z., “The space of human body shapes: reconstruction and
parameterization from range scans”, in ACM ToG, Vol. 22, No. 3, pp. 587-594, 2003.
[15] Ballester, A., Parrilla, E., Uriel, J., Piérola, A., Alemany, S., Nacher, B., González, J. and
González J.C., “3D-Based Resources Fostering the Analysis, Use, and Exploitation of Available
Body Anthropometric Data”, in Proc. of 5th Int.Conf. on 3D Body Scanning Technologies, Lugano,
Switzerland, October 2014. doi:10.15221/14.237
[16] Ballester, A., Valero, M., Nácher, B., Piérola, A., Piqueras, P., Sancho, M., Gargallo, G., González,
J. C., and Alemany S., “3D Body Databases of the Spanish Population and its Application to the
Apparel Industry”, In Proc. of 6th Int. Conf. on 3D Body Scanning Technologies, Lugano,
Switzerland, October 2015. doi:10.15221/15.232
[17] Rother, C., Kolmogorov, V., and Blake, A., “Grabcut: Interactive foreground extraction using
iterated graph cuts”, In ACM Transactions on Graphics, Vol. 23, No. 3, pp. 309-314, 2004
[18] Lu, J. M., and Wang, M. J. J., “The evaluation of scan-derived anthropometric measurements
Instrumentation and Measurement, IEEE Transactions on, 59(8), 2048-2054, 2010.
[19] Dekker, L. D., “3D human body modelling from range data”, PhD thesis, Doctoral dissertation,
University of London, London, United Kingdom, 2000.
[20] Robinette, K. M., & Daanen, H. A., “Precision of the CAESAR scan-extracted measurements”,
Applied Ergonomics, Vol. 37, No. 3, pp. 259-265, 2006.
[21] Structure for iPad, http://structure.io/ (29/09/2016)
[22] Project Tango from Google, https://developers.google.com/tango/ (29/09/2016)
[23] DPI2013-47279-C2-2-R. “Herramientas para la predicción de la talla y el ajuste de ropa infantil a
partir de la reconstrucción 3D del cuerpo y de técnicas 'big data'” (TALLA-ME) Programa Estatal
de Investigación, Desarrollo e Innovación Orientada a los Retos de la Sociedad.
Proceedings of the 7th International Conference on 3D Body Scanning Technologies, Lugano, Switzerland, 30 Nov.-1 Dec. 2016
147
... To select the best features and to estimate the best coefficients for each shoe model, stepwise method was used. IBV has already used logistic regression models to tackle this problem successfully in clothing [19], [20]. In this project, we also used data projection to bring all the observations to a single size by upscaling and downscaling the foot features characterizing each foot, making it work in all available sizes. ...
... The two models that showed a fit tolerance below 70% were models with fit issues that probably affected the size choices of the subjects. These results are in line with similar systems using logistic regressions to recommend sizes [19], [20] and provides better results than just picking the usual size (55-60%) or using the brand's size chat using foot length (45-55%). ...
... MAPE is the mean absolute percentage error(Kim &Kim, 2016). To analyze the error, the Mean Absolute Difference (MAD) was computed(Ballester et al., 2016). MSD is the mean square deviation(Karmaker et al., 2017).The obtained numerical variance is one of the dispersion indicators and shows how far the data are from the average value. ...
Preprint
Full-text available
Ahwaz Hydrometric Station is responsible for controlling surface water resources and the Karoon River near Ahwaz city in southwestern Iran. And the present study aimed to determine the parameters affecting water quality, especially TH and SAR parameters. For this purpose, 39-year old statistical data were collected with 463 samples. To determine the water quality, first the correlation matrix method and statistical analysis were conducted, and then the correlation between them and the accuracy of these methods were checked using the tree model and the K-Nearest Neighbor (K-NN) method. The K-NN method and multivariate regression were compared for water quality characteristics, including SAR. The results indicated that K-NN methods were better than the regression method. In addition, the K-NN method using the effective anion and cation combinations yielded better results of estimating Sodium Absorption Ratio (SAR) and Total hardness (TH). Furthermore, the accuracy of the tree model after estimating TH using SO 4 ²⁻ was more than that of Ca ²⁺ . Moreover, the accuracy of the tree model using the Cl ⁻ data for SAR estimation was higher than that of the Na ⁺ data. In general, according to the APHA standard (2005), river water is in the high-risk and low-alkaline group.
Article
Full-text available
The advances and availability of technologies for the acquisition, registration and analysis of the three-dimensional (3D) shape of human bodies (or body parts) are resulting in the formation of large databases of parameterised meshes from which digital human body models can be derived. Such models can be used for the data-driven reconstruction of parameterised human body shapes from partial information such as one-dimensional (1D) measurements or 2D images. In this paper, we propose a new method for the reconstruction of 3D bodies from images gathered with a smartphone or tablet. Moreover, the method is implemented into a prototype app and tested at different levels through three experimental studies including synthetic models, 1:10 scale figurines and real children. The results demonstrate the feasibility of acquiring reliable anthropometric information easily at home by non-experts. This method and implementation have great potential for their application to the personalisation, size recommendation and virtual try-on simulation of wearable products.
Conference Paper
Full-text available
This paper describes two approaches for estimating human 3D shapes (i.e. full bodies or feet) using a regular smartphone or just entering a set of parameters (e.g. age, gender and self-taken measurements). The proposed approaches are based on data-driven 3D reconstructions, using parameterised shape spaces created from large 3D body or feet databases. The reconstruction algorithm finds the combination of shape parameters that best matches either the silhouettes extracted from the images or the body measurements entered. Despite not being actual body scanners, these solutions are easy-to-use and can provide enough accuracy for applications such as virtual try-on, made-to-measure or size allocation of certain types of wearables. Moreover, they can be distributed to the final consumer or to the points of sale at a really reduced cost (or even for free), thus overcoming the main barriers to the massive spreading of body scanners’ use to e-commerce, retail shops, new production pipelines or new business models. In order to illustrate these technologies, some examples of application to different contexts are provided, namely virtual worlds, e-commerce and personalisation.
Conference Paper
Full-text available
Since the year 2000, many anthropometric surveys have been conducted across the world using 3D body scanning technologies, most of them addressed to the apparel industry. This paper describes the application to the apparel industry of the 3D Spanish surveys (female, male and children) conducted from 2007 to 2015 by IBV gathering over 12.000 individual scans. It also presents tools that will help the apparel manufacturers and retailers to make an effective use of Spanish databases in the design as well as in the labelling of products addressed to the Spanish market and following the forthcoming size designation interval standards (EN 13402). These tools consist of a website providing with the basic anthropometric statistics, two books with the population measurements by age range (one for female and one for male populations), a collection of digital mannequins and a collection of physical mini-mannequins (scale 1/20). Moreover, the access to the 3D databases makes possible to IBV to extend the use of these data for the provision of new consultancy services for clothing companies about how to improve garment design and fitting.
Article
Full-text available
Introduction The access to the 3D representation of people's feet has multiple applications in footwear industry, ranging from custom made shoes to virtual try-on or size allocation. However, there are several barriers that have hindered the massive spreading of 3D scanners as consumer good or as typical in-store appliance: the price is high and the device is too bulky for homes and retail stores; and it requires expertise to achieve a quality scan and to locate anatomical references to get measurements from the 3D object. This paper describes a novel approach for estimating the 3D shape of the foot using a smartphone application. The proposed 3D reconstruction is data-driven, since it uses a parametrised shape space created from a feet database. The algorithm finds the combination of shape parameters that best matches the foot silhouettes extracted from the images.
Conference Paper
Full-text available
Nowadays each clothing company defines its own sizing chart to label the garment. The lack of regulations and the different labelling methods used on each country contribute to have a confusing buying process for the end users in terms of garment size selection. Depending on the brand, the customer selection can vary several sizes. This makes that customers need to try on many sizes to select the suitable one during the buying process. This is one of the main barriers to the growth of the online sales in the fashion market. The high number of returns and their associated costs (e.g. management, logistics and operations) represent an important economic burden for fashion companies. The aim of the this paper is to present a proposal of new methods for the development size selection systems based on a 3D body acquisition process using 3D body reconstruction and a multi-fitting approach to predict garment size.
Conference Paper
Full-text available
Today, there is an increasing availability of human body 3D data and an increasing number of anthropometric owners. This is due to the fact of the progressive conduction of large national surveys using high resolution 3D scanners and due to the increasing number of low-cost technologies for acquiring body shape with electronic consumer devices like webcams, smartphones or Kinect. However, the commercial use and exploitation in industry of digital anthropometric data is still limited to the use of 1D measurements extracted from this vast 3D information. There is a lack of universal resources enabling: to conjointly use and analyse datasets regardless from the source or type of scanning technology used, the flexible measurement extraction beyond pre-defined sets, and the analysis of the information contained in human shapes. This paper presents four software tool solutions aimed at addressing different user profiles and needs regarding the use and exploitation of the increasing number of 3D anthropometric data
Article
Objective: To provide a review of sustainable sizing practices that reduce waste, increase sales, and simultaneously produce safer, better fitting, accommodating products. Background: Sustainable sizing involves a set of methods good for both the environment (sustainable environment) and business (sustainable business). Sustainable sizing methods reduce (1) materials used, (2) the number of sizes or adjustments, and (3) the amount of product unsold or marked down for sale. This reduces waste and cost. The methods can also increase sales by fitting more people in the target market and produce happier, loyal customers with better fitting products. This is a mini-review of methods that result in more sustainable sizing practices. It also reviews and contrasts current statistical and modeling practices that lead to poor fit and sizing. Fit-mapping and the use of cases are two excellent methods suited for creating sustainable sizing, when real people (vs. virtual people) are used. These methods are described and reviewed. Evidence presented supports the view that virtual fitting with simulated people and products is not yet effective. Conclusions: Fit-mapping and cases with real people and actual products result in good design and products that are fit for person, fit for purpose, with good accommodation and comfortable, optimized sizing. While virtual models have been shown to be ineffective for predicting or representing fit, there is an opportunity to improve them by adding fit-mapping data to the models. This will require saving fit data, product data, anthropometry, and demographics in a standardized manner. For this success to extend to the wider design community, the development of a standardized method of data collection for fit-mapping with a globally shared fit-map database is needed. It will enable the world community to build knowledge of fit and accommodation and generate effective virtual fitting for the future. Application: A standardized method of data collection that tests products' fit methodically and quantitatively will increase our predictive power to determine fit and accommodation, thereby facilitating improved, effective design. These methods apply to all products people wear, use, or occupy.
Thesis
This thesis describes the design, implementation and application of an integrated and fully automated system for interpreting whole-body range data. The system is shown to be capable of generating complete surface models of human bodies, and robustly extracting anatomical features for anthropometry, with minimal intrusion on the subject. The ability to automate this process has enormous potential for personalised digital models in medicine, ergonomics, design and manufacture and for populating virtual environments. The techniques developed within this thesis now form the basis of a commercial product. However, the technical difficulties are considerable. Human bodies are highly varied and many of the features of interest are extremely subtle. The underlying range data is typically noisy and is sparse at occluded areas. In addressing these problems this thesis makes five main research contributions. Firstly, the thesis describes the design, implementation and testing of the whole integrated and automated system from scratch, starting at the image capture hardware. At each stage the tradeoffs between performance criteria are discussed, and experiments are described to test the processes developed. Secondly, a combined data-driven and model-based approach is described and implemented, for surface reconstruction from the raw data. This method addresses the whole body surface, including areas where body segments touch, and other occluded areas. The third contribution is a library of operators, designed specifically for shape description and measurement of the human body. The library provides high-level relational attributes, an "electronic tape measure" to extract linear and curvilinear measurements,as well as low-level shape information, such as curvature. Application of the library is demonstrated by building a large set of detectors to find anthropometric features, based on the ISO 8559 specification. Output is compared against traditional manual measurements and a detailed analysis is presented. The discrepancy between these sets of data is only a few per cent on most dimensions, and the system's reproducibility is shown to be similar to that of skilled manual measurers. The final contribution is that the mesh models and anthropometric features, produced by the system, have been used as a starting point to facilitate other research, Such as registration of multiple body images,draping clothing and advanced surface modelling techniques.
Article
Achieving well fitting garments matters to consumers and, therefore, to product development teams, garment manufacturers and fashion retailers when creating clothing that fits and functions both for individuals and for a retailer's target populations. New tools and software for body scanning and product development enhance the ways that sizing and fitting can be addressed; they provide improved methods for classifying and analysing the human body and new ways of garment prototyping through virtual product development. Recent technological developments place a growing demand on product development teams to reconsider their approach to prototyping, sizing and fitting. Significant, related changes are also being made in the fashion retail environment, including innovations in virtual fit to enable consumers to engage with fit online. For best effect in the short term, such advances need to relate well to existing manufacturing practices and to the methods that have, over many years, become embedded by practitioners into the processes involved in clothing product development and those used for establishing garment fit. The high rate of technological advance, however, places an urgent need on practitioners to change; established principles of pattern theory need to be recognised explicitly and followed consistently, otherwise, new techniques for developing and assessing products will not be able to be fully exploited. Practitioners will be pressed to adopt more data-rational approaches to product development, including adopting engineering principles into the practice of clothing product development. For example, comparisons made between the traditional two-dimensional garment pattern and the three-dimensional environment accessible through 3-D body scanning technology, provide both the stimulus and the data required to support a re-examination of how the measurements required for clothing product development should be defined. This should be coupled with a more explicit recognition of ease as a factor requiring quantification within clothing engineering. New methods of categorising the body in terms of its form also allow recognition of the restrictions of proportional theories in pattern construction; they afford promising opportunities for advancing the practices of sizing and fitting in clothing product development.
Article
An anthropometric study was performed on 103 women 19 to 50 years of age. Each subject measured 19 dimensions of her own body and then performed similar measurements on a random partner using a conventional tape measure with 1/16 inch (1.6-mm) precision. The experimenter measured all subjects using a laboratory-grade anthropometer and tape measure with 1.0-mm precision as a standard. Subjects' measurements of different body dimensions had significantly different average errors, ranging from -4.54 cm to +6.15 cm. Hip circumference, which is often utilized as a key dimension for garment sizing, was significantly undermeasured (M = -4.54 cm). Measurements having the least error included waist, bust, and neck circumferences and shoulder and waist heights. Subjects' measurements of their own bodies had greater absolute error (M = 4.10 cm, SD = 6.06 cm) than those of their partners (M == 3.34 cm, SD = 4.86 cm). Although subjects used a tape measure having 1.6-mm precision, 97% reported measurements using only 1/4-inch (6.4-mm) precision or less. Subjects overestimated their own stature by an average error of 0.68 cm (SD = 4.02 cm) and an absolute error of 2.26 cm (SD = 3.39 cm).