Skip to main content

Intelligent 3D Design and Simulation System for the Clothing Industry

Objective


During the project, the consortium has carried out a great research effort to achieve the objectives of the project. Some industrial prototypes have been built and successfully evaluated. The consortium has learned very much along the project, and the industrial prototypes are very close to be converted in final products. Many achievements have been done, but new challenges have appeared and another problems stay.
A 3D modeller oriented to the garment design process has been developed. The 3D modeller allows to define garments pieces directly on a 3D mannequin; to develop the 3D panels to get the 2D patterns; to create and to edit 3D lines on the 3D garment: darts, seamlines, cutting lines; and to export the garment towards the fabric simulator. Some garment models have been created and the physical prototypes have been successfully built using the output 2D patterns. The modeller have been developed by the Univ. of Valenciennes, and it is an excellent example of the potential of 3D technology on the garment industry.

We have developed a 3D Simulator that allows the modellist to simulate the behaviour of the designed garment on a mannequin according to physical properties of the fabric; therefore it permits to verify the garment style on a virtual mannequin comparing different types of fabric (e.g., cotton, silk). The system manages interactions with the environment taking into account external forces (e.g., gravity), collisions (e.g., with the mannequin) and auto-collisions. The prototype has been experimented both using data generated by the 3D modeller and by the 2D-3D mapping module developed by IT vendors. The results are encouraging and the system is more suitable to the garment design problem than commercial systems more oriented to the animation industry. The research effort has been considerable, although the main limitation still remains the computational cost that depends on garment complexity. Better performances, from computational point of view, can be obtained implementing the prototype on a parallel platform.
Both the 3D modeller and the 3D Simulator have been presented at an international workshop on geometric modelling, GEO7, held in Parma, 2-3 October 2000 and at International Seminar on Modelling and Virtuality held at Treviso, 16-17 November 2000.
Another 3D basic modeller has been developed to evaluate others flattening algorithms and to develop a parametric mannequin according to the garment industry. This prototype allows adjusting a parametric mannequin according to size measurements values and looseness values. The parametric mannequin defines two surfaces: the body surface and the garment surface. The garment surface is an abstract surface relative to the body surface. The garment surface is defined by the looseness table, and it represents the space where the garment is placed.

The Abstract Garment Surface has allowed us to introduce the Universal Pattern Definition representation, which allows defining a valid pattern on any body shape, because this representation separates the geometric description of the pattern (topology) from the size measurements of the body. The modeller includes other flattening algorithm, which have been evaluated, getting good results. The grading of the patterns is automatically computed when the user updates the size or looseness values. We have created and built some garment prototypes with this tool, obtaining encouraging results. INVESTRONICA introduced this modeller prototype during the last IMB'2000 exhibition at Cologne, getting and excellent feedback of the public. INVESTRONICA has the intention to develop a commercial product based in this technology, including all the 3D edition functionality necessary to close the product. Following this direction, INVESTRONICA has introduced to its customer base and dealers this product like the core of a new line of 3D products, which be coming during the next year.

A 2D-3D mapping tool was developed, which allows to visualize a complete 3D garment on a mannequin, assembled from its 2D patterns. The tool allows to sewn and to place the 2D patterns of a garment on a 3D virtual mannequin, giving a virtual try-on of the garment model. This prototype is the result of the join effort of LECTRA and INVESTRONICA \cite {Yepes99}, and it has been introduced by LECTRA during the last IMB'2000 exhibition at Cologne, getting and excellent feedback of the public and customers. LECTRA is working to achieve a final product based in this technology. LECTRA also has the intention to close the 2D-3D mapping product, and they also have introduced this product to its customer base and dealers.

We have learned so much about the 3D representation of the human body, and the information necessary to import these data in a 3D garment design process. TELMAT has work hardly to improve and to adapt the SYMCAD system to the requirements of the 3D garment design process. The 3D raw data of the human body obtained with SYMCAD is excellent and enough to develop a Made To Shape design tool like the proposed in the 3D modeller. The main difficulties have been to define how to export and to represent these data, such that a 3D modeller could use. This situation is consequence of the difficulty to achieve a valid geometric representation for the mannequins, and the dependency between the mannequin model and 3D data outcoming from the scanner. Only when the modellers and parametric mannequin were very advanced, we had an approximate idea about the problem. Finally, the results obtained have been satisfactory and very encouraging. We could import these data in the modeller prototype of INVESTRONICA, and to build a simple garment. TELMAT and INVESTRONICA are working to produce a standard exchange format to achieve the real mass application of the 3D MadeToShape technology. This project has allowed gaining an invaluable experience in this sense and we think that in a short time period we can offer to the market real MadeToShape (MToM) 3D garment design.
Garment companies depend more and more for their
competitive strength on the possibility of shortening the
design cycle of new styles and reducing the trial and
error process which has been the traditional approach for
garment design.
The industrial garment design process is composed of:
styling (design of stylised draft representing the
ganment), dress design (interpretation of the stylised
drafts to produce a ganment prototype), pattern handling
(grading, industrialisation, and control of 2D patterns).
Most of the individual manufacturing operations are
covered by software tools allowing a great saving in time
and material (fabric), for example:

2D painting software's which provide free drawing
fashion oriented tools
2D handling software's which allow pattern
industrialisation
Interactive Grading
Marker making

Each of these tools covers a particular step of the
ganment design process but coherence is not ensured
because of the lack of software for the whole process of
garment design.
In such a context, the main goal of the project is to
propose a very innovative approach where the design of a
new ganment can be carried out directly on 3 D and the 2D
patterns are automatically generated.

The proposed system will allow the designer to work
easily with virtual 3 D frames on which 2D pieces of
fabric can be laid (considering their mechanical
properties) in the preferred design. The model can then
be rotated or a split screen used to view it from
different angles simultaneously. Lighting, drape and
gravity are simulated realistically. Pieces of the
ganment can be modified on the model.

Once the designer is satisfied, the system can
automatically generate 2D patterns whose coherence is
guaranteed. Grading, the process of making patterns for
different sizes of ganment, is automatic. It will also be
possible to simulate the actual mechanical behaviour of
the cloth under contact with the human body, gravity and
other forces.
The approach ensures the correct fitness of the garment
being designed and it will reduce by three times the
current trial and error cycle of new designs on human
models. The costs of industrialisation and control of 2D
patterns will be improved by at least 30%. by using this
approach.
The grading process which accounts for 20% to 30% of the
design costs can be drastically reduced because the
grading will be done by changing the body measurements of
the 3D electronic mannequin.

The technology to be developed in the project combined
with 3D measurement devices will also be used to
automatically generate made to measure garments so that
the customer can immediately see the fitness and the look
of the garment.

The consortium comprises two leading manufacturers of
CAD/CAM systems for the garment industry, two research
institutes with deep knowledge in 3D modelling and cloth
simulation, an industrial company specialising in
designing advanced technology systems for Made Measure of
garments and three different ganment manufacturers
(women, men, children) which will help with
specification, validation and assessment of the
prototype.

Funding Scheme

CSC - Cost-sharing contracts

Coordinator

Investronica Sistemas SA
Address
62,Tomas Breton 62
28045 Madrid
Spain

Participants (7)

Confecciones Mayoral SA
Spain
Address
1,Camino De Los Prados
29006 Malaga
Defence Clothing and Textiles Agency
United Kingdom
Address
Monxton Road
SP11 8HT Hampshire
Gruppo Finanziario Tessile SpA
Italy
Address
Corso Emilia 6
10152 Torino
Lectra Systèmes SA
France
Address
Chemin De Marticot
33610 Cestas
Telmat Informatique Eurl
France
Address
6,Rue De L'industrie
68360 Soultz
UNIVERSITÀ DEGLI STUDI DI PARMA
Italy
Address
Viale Delle Scienze
43100 Parma
UNIVERSITÉ DE VALENCIENNES ET DU HAINA UT-CAMBRÉSIS
France
Address
Le Mont Houy
59304 Valenciennes