Nowadays, virtual try-on is an irreplaceable technology in fashion industry, so it is very important to prove virtual try-on matching with the real garments. Therefore, the aim of this research was to compare garment fit using virtual try-on and scanning technologies. For this reason, garment visual appearance and distance ease between straight fit dress and mannequin in respect to fabrics properties were investigated. Women mannequins in different sizes were scanned by 3D scanner VITUS Smart XXL without and with the real straight fit dresses made from five different woven fabrics. Fabrics mechanical properties were defined by KES-F. Scanned mannequins were covered with the same size and fabric virtual dresses by Modaris 3D (CAD Lectra). Distance ease of virtual and scanned garments was compared in bust and waist cross-sections. It was defined that distance ease values at bust girth of real and virtual dresses differed till 29.9 % (1.16 cm), while at waist varied from 7.3 % (0.51 cm) to 47.3 % (4.30 cm) because of wrinkles in this area. Generally, appearance of the virtual dresses was similar to real dresses with some differences in garment shape fluency, however by increasing of the mannequin size, similarities decreased. It was assumed that very high shear rigidity G could not be very well reflected in 3D CAD system, therefore differences between virtual and real dresses appearance occurred. The general appearance and form of bust and waist cross-sections of virtual dresses with fabric 03 had less similarities comparing with real dresses due to high G value. So, comparative study showed that the accuracy of virtual try-on was quite useful comparing to real garments, if shear rigidity of fabrics was lower than 1.6 N/m*º and tensile strain in warp direction was higher than 1.80 %.
A scheme for functional tolerancing: A product family in 3D CAD system
