CC BY
42этом помогают специально разрабатываемые каждым брендом приложения для смартфонов. В работу дизайнера прочно входит сотрудничество с искусственным интеллектом.
Список литературы
1. Agins T. The End of Fashion: The Mass Marketing of the Clothing Busines. New York.: William and Morrow, 1999.
2. Васильев А. Формула моды: Тайны прошлого, тренды настоящего, взгляд в будущее. М: БОМБОРА, 2021.
3. Голуб А. Искусственный интеллект для моды. Минск: Дискурс, 2019.
4. Клайн Э. Осознанный гардероб. Как выглядеть стильно и спасти планету. М.: ОДРИ, 2020.
DOI
3D TECHNOLOGIE IN FASHION BUSINESS. INCREASE THE CUSTOMS SATISFACTIONS
E.S. Likhacheva1, P.A. Egarmin1, Prof. Vladan Koncar 2
'Saint-Petersburg State University of Industrial Technologies and Design, 156236, Russia, Saint-Petersburg
2Ensait School of Engineering, 59056, France, Roubaix
The article presents a broad overview of the use of 3D modeling technology, 3D scanning of the human body, virtual fitting. The authors analyzed user preferences when buying goods online. The influence of 3D technologies for the fashion industry was studied, it was found that 3D technologies allow online stores to reduce the number of returns and increase the average check. Based on these studies, a system of recommendations is presented on how to increase customer satisfaction by giving them the opportunity to change the design of the product, and how to automatically adapt the patterns so that they are ideally suited for any client. Based on a professional apparel knowledge base, virtual fitting experience, and return statistics, the authors try to find a solution to make the 3D model and the real product as similar as possible.
Digital clothing has become a reality of modern fashion, and is already used by both consumers and businesses - designers, brands and fashion retailers. A promising option is to use digital clothing not as a final product, but as a tool for preliminary demonstration of goods: the possibility ofpresenting a future collection for pre-order and reselling goods before the release of a real collection. Fashion retail can use digital models at the stage of preparing the collection - before it is finished, as a preview for pre-order. Another possible, but not yet implemented, option for using virtual clothing in fashion retail is tailoring on demand, when the buyer gets the opportunity to fully adapt the chosen model for himself and try on the result in advance in the virtual space. World leaders in fashion are moving towards 3D, towards mannequins adapted to people in order to be able to create clothes on demand. With a 3D twin of clothing, production costs are significantly reduced, for example, during the development of a 3D model, instead of two or three stitched samples, only one will be needed. Moreover, companies will be able to produce a range tailored to consumer demand in optimal quantities. Some foreign brands present a new collection immediately in 3D, offering users to pre-order, thus assessing the potential of each assortment model for the target audience, and then form the final assortment matrix for mass production in optimal quantitative proportions, significantly reducing their costs. 3D design gives limitless possibilities for customization - clothes are fully adjusted to the buyer - fit perfectly and can be changed in any detail to the taste and style of the owner.
Key words: Customisation, codesign, personalisation, AR, VR, 3D, virtual try-on, fit evaluation, return rates, online sales, avatar, scan.
УДК 7.06
3D ТЕХНОЛОГИИ В МОДНОМ БИЗНЕСЕ.
КАК УВЕЛИЧИТЬ УДОВЛЕТВОРЕННОСТЬ ПОКУПАТЕЛЕЙ
Е.С. Лихачева1, П.А. Егармин1, В. Кончар2
1 Санкт-Петербургский государственный университет промышленных технологий и дизайна, 156236, Россия, Санкт-Петербург
2Высшая национальная школа искусств и промышленного текстиля, 156236, Франция, Рубе
В статье представлен широкий обзор использования технологии 3D моделирования, 3D сканирования тела человека, виртуальной примерки. Авторами проведен анализ пользовательских предпочтений при покупке товаров в онлайн среде. Было исследовано влияние 3D технологий для модной
индустрии, выявлено, что 3D технологии позволяют интернет-магазинам уменьшить количество возвратов и увеличить средний чек. На базе этих исследований представлена система рекомендаций как повысить удовлетворенность клиентов, предоставив им возможность изменить дизайн изделия, и как автоматически адаптировать лекала, чтобы они идеально подходили для любого клиента. Основываясь на профессиональной базе знаний об одежде, опыте виртуальной примерки и статистике возвратов, авторы пытаются найти решение, как сделать 3D-модель и реальный товар наиболее похожими.
Цифровая одежда стала реальностью современной моды, и уже используется как рядовыми потребителями, так и бизнесом: дизайнерами, рендами и fashion-ритейлерами. Перспективным видится вариант использования цифровой одежды не как конечного продукта, а как инструмента для предварительной демонстрации товара: возможности презентации будущей коллекции для предзаказа и перепродажи товаров перед выпуском реальной коллекции. Представители fashion-ритейла могут использовать цифровые модели на этапе подготовки коллекции (до отшива) как превью для предзаказа. Еще один возможный, но пока не реализованный вариант использования виртуальной одежды в fashion-ритейле - пошив одежды on demand, когда покупатель получает возможность полностью адаптировать выбранную модель под себя и заранее примерить полученный результат в виртуальном пространстве. Мировые лидеры моды идут в сторону 3D, в сторону манекенов, адаптированных под людей, чтобы иметь возможность создавать одежду on demand. С 3D-двойником одежды затраты на производство существенно сокращаются, например, во время отработки 3D-модели вместо двух или трех сшитых образцов понадобится только один. Более того, компании смогут производить ассортимент, адаптированный к потребительскому спросу, в оптимальном количестве. Некоторые зарубежные бренды представляют новую коллекцию сразу в 3D, предлагая пользователям сделать предзаказ, таким образом оценивая потенциал каждой модели ассортимента на целевой аудитории, а затем уже формируют окончательную матрицу ассортимента для массового производства в оптимальных количественных пропорциях, существенно сокращая свои затраты. 3D дизайн дает безграничные возможности для кастомизации: одежда полностью подстраивается под покупателя, идеально садится и может быть изменена в любых деталях под вкус и стиль владельца.
Ключевые слова: кастомизация, кодизайн, персонализация, AR, VR, 3D, виртуальная примерка, оценка посадки, процент возвратов, онлайн-продажи, аватар, скан.
Introduction
With increasing online business, there are a lot of customers buying clothes from e-stores. There is a growth of the online business due to several reasons:
1. Current situation with COVID. According to McKinsey, 20-30% of businesses moved online during the pandemic's peak. Worldwide, 71 percent of fashion executives expect their online businesses to grow by more than 20 percent in 2021, which is close to the record-breaking pace set in 2020 [1].
Shop online
Won't shop online
2020
2021-22 (projected)
75 25
85 15
Fig. 1. Current and projected share who shops online, %
Develop technology. "We'll experience more technological progress in the coming decade than we did in the preceding 100 years put together", says McKinsey [3]. Key technologies for the fashion industry [2] (fig. 2.).
Fig. 2. Key technologies for the fashion industry
There are a lot of advantages to online business exposed below:
1. Online stores may be opened on Saturday and Sunday during the night, 24 hours, 7 days in a week.
2. People who build websites don't have to invest in big shops. They do not have to build a big malls because it is very expensive. Moreover there is no rent to be paied.
3. It is possible to increase the number of possible customers because online stores are visible worldwide. Everyone can connect and the potential base of customers is large, if the proper marketing investments are realized
4. Designers are able to send their products to customers in different countries as the e shop is available in English or in French, or in different other languages.
There are a lot of different reasons why people buy online. However, there are a number of disadvantages.
The most important drawback of online sales is the customers' satisfaction issues because when the real products are received they may be quite different from the virtual ones seen on the mannequins on the computer or smartphone screen. These problems generate important return rates of the real products, which are very costly for the companies involved in the virtual merchandising.
Fashion products are the top-selling category in online retail. It accounts for about $596.1 billion in 2020, worldwide, and continuously escalating (fig. 3) according to statista.
What is the virtual product?
A virtual product (also known as digital goods) is a non-tangible item that can be downloaded or traded in a virtual economy, such as in e-commerce sites, online groups and marketplaces. There is a profound growth of garment purchases through the Internet. In this context, fit evaluation of virtual garment try-on is vital in the clothing industry [5]. The fit of garments has serious implications for a fashion retailer because ill-fitting garments are directly related to product return rates. The evaluation of garment fit without the physical participation of customers and designers is very useful for online clothing shoppers.
Dresses Dresses Dresses
mm ■■[ mm
Fig. 4. 3D dresses
Virtual product could be different in following situations:
1. 2D models - photo.
2. 3D models.
3. Video, fashion show, catwalk.
4. 3D models with codesign. It means that a customer can modify some parts or some combinations of colors or whatever and therefore participate in the design personalization. (Fig.4)
Also, if a photo of the virtual product is blur and/or low resolution the real received products have a chance to be very disappointing. The size is another issue, as the customers are sometimes very optimistics.
The main reasons why the real product can be different from the virtual product are:
1. Color mismatch;
2. Fabric quality;
3. Sewing quality;
4. Size, shape and poor fit.
The color depends on the personal device. The color displayed on a screen $ and real fabric color are never the same. If fabric and sewing are not good the 3D e-shop approach will not fix this issue and the customers will not be satisfied. This may be solved by the quality process and better control.
The size and the shape issues may be solved by using a 3D model of the customer and by adapting the patterns to the body scan. In this case the size and the shape will be rather good for the customer who purchased this product. The real product may be made as similar as possible to the virtual one because of the patterns perfectly adapted to the person. In that way a real product will not be very different from the virtual.
With the body scan and codesign approach use the customers will be more satisfied and this could solve the problem of the high return rate:
The process including previously announced improved tool supposed to increase the customers' satisfaction and to decrease the return rate are defined below:
1. Adapt patterns for individual size : To perform this adapting it is necessary to develop a website able to use the customers measurements issued from her or his body scan and an appropriate software aiing at the modification of the basic patterns to make the final product perfectly fitted to the customer.
2. Give a possibility to customers to personalize their clothing: to achieve this mass customization and co design concept it is necessary to offer a possibility to customers to personalize their products by adding some embroidery or by changing a set of colors etc. In that way, we believe the customers will better identify that the purchased clothing is unique and their. Therefore, they will be less kinn to send them back by decreasing the return rate.
Scan measurements generated from a body scan focus on the "tailor" principle but in an automated way. It is done by software and may be scale up to a large number of customers. In fact, the proposed approach will just generalize the tailor measurements on a large scale. The problem that may occur because of a lack of body scanners may be solved by their implementation in malls with a small cost to get a body scan, or even by the use of the apps for smartphones able to generate the body scan from several photos, but with improvable precision.
Usually customers are ready to pay a little bit more (from 10% to 30%) for being satisfied because they will have something that is only for them. As customers are not designers we don't have to give them too many possibilities, because they could make something not very attractive, but on the other side it is difficult to define a subjective feel of the attractiveness. In our opinion it woul be better to give a limited set of modifications to the customers to keep the original design as much as possible. From the psychological point of view it may be sufficient to make them happy with the product and to avoid its return. The problem is still
1. How to make more profit?
2. How to increase customer satisfaction?
3. How to decrease this return rate?
4. How to make this process automated?
The answer to the questions above is probably in the development of reliable software able to connect the previously mentioned principles with the virtual try on tool. can perform some co design and if they are happy with the design they try vrtually their new clothing and order them. As the software already had modified patterns, a designer does not have to prepare the patterns again. Designer just sends those modified patterns of the cutting tables and realizes the sewing. The customer who will receive this garment will have the same real garment as similar as possible to the 3D virtual model of the garment. To automate this process is however very costly, but may increase the customers satisfaction and decrease the return rate. The designers will be able to produce and cut much more patterns implying the need for the more intensive production (fig.5).
Fig. 5. Plan for making the real and virtual product
How to make a good fit
A number of virtual try-on programs, such as Clo 3D, Lectra 3D Prototype, OptiTex and V-Stitcher 3D, are available on the market for garment fit evaluation. These 3D virtual try-on software systems follow similar principles, i.e. showing virtual garment static and dynamic performance from identified human morphological and fabric properties and their interactions. Defining this performance involves the use of complex mechanical and geometric modeling and simulation techniques, such as finite elements [3].
Codesign and personalization
Mass customization (according to Wikipedia), in marketing, manufacturing, call centres, and management, is the use of flexible computer-aided manufacturing systems to produce custom output. Such systems combine the low unit costs of mass production processes with the flexibility of individual customization.
We used to think that it is only two kind of clothes: Ready-to-wear and Made-to Measure. Ready-to-wear we can buy cheap price and standard size. Made-to-Measure is a long process and cost expensive but fit perfectly. Mass customization helps to combine these two process in one and give more satisfaction to custom and help productions to automatized system.
Ready-to-Wear:
1. Not expensive.
2. Available.
3. Well-fit.
Made-to-Measure:
1. Well-fit.
2. Expensive.
3. Experience.
Mass customization = Ready-to-Wear + Made-to-Measure
1. Not expensive.
2. Available.
3. Well-fit.
4. Automatical.
Softwear for customization is Vivaty, SketchUp or another 3D softwear. They help to change some thing in garment and make it possible to choose what customers really want. (Fig.6)
Fig. 6. Customised dress 3D model and real garment Virtual try-on before buying
For the design process, an adaptive women body mannequin which was previously designed using Design Concept software with the relevant data has been used. The woman's mannequin was attained using a 3D scanner and the software scanWorx from the Human Solutions Company. After attaining the scanned body shape, its data was imported to software, called Rapidform. This software helps us to edit and correct the defects of the imported 3D meshed object. Finally, using the 3D Design Concept software, the 3D surface of the woman's body shape as shown in Figure 7 can be modeled. The operation of 3D scanning permits one to directly obtain the 3D body shape, on which the process used for analyses and modifies the body shape. [6] Modify patterns to avatar (fig 8) and then use the cutting mushine.
Fig 7. Body scan
Fig. 8. Patterns in Lectra Modaris 139
If the patterns have been made for the customer and customer have personalized e-garment. The customer can check on avatar outlooks how it fits. Using Lectra Modaris (fig. 9) or Clo 3D we can check the pressure intention and if it is too small or too big we can change size before making a real garment.
Conclude discussion
In this paper, we proposed an idea on how to satisfy customs and decrease the return rate for online sales. For perfectly fit clothes better use try-on with 3D software and get measurements from scanner and avatars. Use programs for adapting patterns automated. The important step is to let the customers personalize their product to make them unique. The perfect fit and personalization are suitable for garment e-shopping and for garment products development.
In practice, this method is known, but nobody did it. The reason is that new software is costly for companies. But the brand needs only once to buy it and then it is possible to save time and money while working.
References
1. Six vectors of success in online fashion. - 2021. - Article https://www.mckinsey.com/business-functions/growth-marketing-and-sales/our-insights/six-vectors-of-success-in-online-fashion.
2. The State of Fashion Technology. - 2022 - PP. 13.
3. Sean Fleming, "Top 10 tech trends that will shape the coming decade, according to McKinsey," World Economic Forum. - 2021. - https://www.weforum.org/agenda/2021/10/technology-trends-top-10-mckinsey/.
4. Fit evaluation of virtual garment try-on by learning from digital pressure data Kaixuan Liua, Xianyi Zeng, Pascal Bruniaux, Jianping Wang, Edwin Kamalha, Xuyuan Tao. - 2020. -https://www.researchgate.net/publication/325172567_Garment_Fit_Evaluation_Using_Machine_Learning_Techn ology.
5. Garment Fit Evaluation Using Machine Learning Technology. - 2018.-https://www.researchgate.net/publication/325172567_Garment_Fit_Evaluation_Using_Machine_Learning_Techn ology.
6. Customizations of women bullet-proof jacket through 3D design process. - 2020. - PP. 3.
