Efficient Representation of Garment Fit with Elastane Fibers Across Yoga Poses in 3D Fashion Design Software: A Preliminary Study Using CLO 3D Software

Abstract

With the growing adoption of CLO 3D in the fashion industry and educational settings, the need for accurate material representation and fit simulation in virtual environments is increasing. This study aimed to evaluate whether CLO 3D, without the aid of physical samples, can reliably simulate clothing pressure for compression wear made from different materials. Unlike previous CLO 3D studies that focused on design or pattern accuracy, this study critically examined material-specific simulation limitations and proposed technical enhancements. Two types of leggings with varying spandex content were tested across five yoga poses using the CLO 3D software(version 2024.2.214). The results showed that CLO 3D did not detect differences in clothing pressure caused by variations in spandex content. Furthermore, the pressure values remained constant across different poses for both fabrics, failing to reflect realistic mechanical differences. The highest total clothing pressure was recorded in the Lunge pose (277.02 kPa), and the lowest in the Plow pose (241.37 kPa). These findings suggest that the current simulation engine lacks sensitivity to fabric-specific mechanical properties and movement-based variation. To address these limitations, this study proposes five optimization functions for CLO 3D, including material property input, technical textile databases, environmental condition settings, AI-based comfort prediction, and data management tools. These proposals are expected to strengthen the scientific validity, functional realism, and user-centered applicability of CLO 3D in designing sportswear, medical compression garments, and customized apparel.

1. Introduction

CLO 3D (version 2024.2.214, CLO Virtual Fashion Inc., Seoul, Republic of Korea) is a three-dimensional fashion design software that creates virtual true-to-life garment visualization using cutting-edge simulation technologies. The program is a commercial, closed-source platform that employs a physics-based particle–spring algorithm and position-based dynamics to replicate fabric drape, deformation, and collision with avatars in real time [1,2]. Currently, CLO 3D is utilized in various ways across the fashion industry, enabling designers, brands, and manufacturers to maximize efficiency and creativity throughout the design, production, and marketing processes. Today, CLO 3D is primarily used to inspect designs using digital simulations before producing physical samples. CLO 3D replicates physical properties similar to those of actual fabrics in a virtual environment, allowing designers to check the fabric, fit, and silhouette of the final product before production and easily make modifications within the program without repetitive physical sampling. The software can be integrated with existing computer-aided design (CAD) systems to accelerate pattern creation and reduce errors. In particular, minimized physical sampling enabled by digital simulations helps reduce production time and costs, thereby maximizing the production efficiency of the clothing industry, contributing to reducing resource waste and carbon emissions, and addressing sustainability issues in the fashion industry [3].

CLO 3D can also be effectively utilized in marketing processes. Recently, there have been showcases of garments generated by CLO 3D on virtual runways (e.g., Balenciaga 2021 F/W Digital Collection, Marni 2022 S/S Metaverse Collection, The Fabricant–Digital fashion house from The Netherlands [4]), has after receiving significant attention as a non-face-to-face marketing tool in situations such as the COVID-19 pandemic. Consumers can experience digital garments created with CLO 3D using a 360-degree view function before making purchase decisions. Furthermore, avatar customization has received much interest through the social-based metaverse platforms increasingly used by the younger generations. Consequently, CLO 3D-generated clothing for avatars is being used in digital content fields such as gaming and augmented reality (AR)/virtual reality (VR), transcending the boundaries of fashion and converging with new industries through cross-platform applications [5].

Thus, CLO 3D has established itself as a core technology in the digital fashion industry, going beyond being a mere clothing production tool, driving innovative transformations across the entire process from design and production to marketing and sustainability, and playing a crucial role in the digital transformation and eco-friendly vision of the apparel industry. Prior research on CLO 3D can be broadly classified into three areas: fashion design using CLO 3D, its application as an educational tool, and the optimization of apparel production processes.

First, regarding fashion design research using CLO 3D, Nguyen and Nguyen (2022) [6] have conducted a statistical analysis of the anthropometric dimensions of 378 men aged 30–60 years living in southern Vietnam. They designed and evaluated men’s Ao dai using CLO 3D. Qi (2023) [7] has studied Macau native Portuguese women’s costumes from cultural and historical perspectives and proposed new designs using CLO 3D. Yu and Zhu (2024) [8] have demonstrated that the digital technology of CLO 3D enables a rapid virtual simulation of the traditional clothing of Mazu, known as the guardian deity of the sea.

Second, regarding research using CLO 3D as an educational tool, Essa et al. (2024) [9] have measured the effectiveness of learning the tools and functions of the CLO 3D software among 10 female clothing students at Cairo University, Egypt. The results indicate that the participants’ creative thinking and learning skills improved. Salakhov et al. (2024) [10] have dealt with the prospects and possibilities of the CLO 3D program for the artistic design of clothes. The basic techniques and tools used in designing and modeling the patterns of clothes were analyzed. Widiyawati et al. (2024) [11] have revealed that CLO 3D significantly enhances students’ creativity, leading to a 23% increase in design diversity and reducing material waste by 35–40%, thereby promoting sustainable design practices.

Third, regarding the research on optimizing the garment production process using CLO 3D, Nam and Kim (2021) [12] have verified the similarity and accuracy between virtual and actual fitting in CLO 3D for necklines, which are clothing details that play a crucial role in forming individual impressions and images. Choi (2022) [13] has examined the current status of 3D virtual simulation systems and their impact on companies, discourses on the uncanny valley surrounding avatars, changes in fashion design process derived from 3D virtual simulation systems, co-design and customization in online platforms, and prospects of 3D virtual garments in the fashion and gaming industries. Mohamed et al. (2023) [14] identify that there is no significant difference between real and virtual methods and outline the efficiency of the 3D garment software and the reduction in the time and waste of fabric during the garment manufacturing process.

These prior studies have mainly focused on design and esthetic aspects, with no research on fabric and wearer comfort. The comfort of a garment in relation to the wearer’s body movements is a crucial characteristic of all types of clothing. Particularly in sportswear, where a wide range of motion is required, the stretchability of the fabric and the resulting clothing pressure are important issues. Among various sports, yoga, which involves large and diverse body movements, requires compression wear, which applies pressure on the wearer’s body to provide specific functional benefits [15]. The popularity of leggings among compression wear has been continuously increasing along with the athleisure boom; in the past three years, prior research has focused on studying legging design [16,17], satisfaction [18,19] and clothing pressure [20,21]. Recent studies have also addressed compression garments using 3D prototyping and user testing [22,23]. However, no study has used CLO 3D to measure the clothing pressure of leggings based on different yoga poses and spandex content.

Spandex (also known as elastane) is a segmented polyurethane fiber, generally composed of at least 85% polyurethane, blended with other minor components such as polyester or polyether, according to ISO 2076 [24]. It is a synthetic fiber that plays a crucial role in the stretchability and fit of leggings. The higher the spandex content, the softer and more stretchable the fabric, whereas a lower spandex content provides better durability and slight pressure. Spandex generally has low breathability and hygroscopicity, and as its content increases, these properties can degrade; thus, it is common to blend it with nylon to compensate for breathability when producing leggings [25]. Therefore, it is important to select a suitable material based on desired wearability and activity type. For example, among compression wear fabrics made of nylon and spandex, a fabric composed of 63% nylon and 37% spandex has been reported to have a stronger compressive strength than a fabric made of 75% nylon and 25% spandex [26]. Because the overall wearability of compression wear can vary depending on differences in various properties, such as stretchability and recovery, when designing compression wear, the selection of a suitable material (or fabric) is more important than the design.

Clothing pressure, that is, the pressure exerted by the fabric on the body, can be measured in CLO 3D and applied in the design of clothing. Liu et al. (2018) [27] have found that clothing pressure data from 3D virtual fitting software have predictive accuracy for the assessment of garment (more specifically, pants) fitness. However, they used a limited number of fabrics (i.e., cotton and polyester fabrics) and wearers’ postures, and relatively simple postures with a limited range of motion, such as sitting, walking, and raising the arms. Therefore, their study does not comprehensively reflect the wearability and fit characteristics for various physical activities [28].

Accordingly, this study aims to verify whether it is possible to accurately and sufficiently represent the material aspects of functional clothing for physical activity using only the CLO 3D virtual fitting system. Specifically, we analyze the differences in clothing pressure according to spandex content in the compression wear when performing various yoga poses based on 3D virtual fitting. This study evaluates the material representation capabilities of the current 3D virtual fitting system and proposes better ways to utilize and optimize the software in fashion design.

2. Materials and Methods

2.1. Virtual Try-On Garment Creation

In this study, three-dimensional virtual try-on garments using a pattern based on the small (S) size leggings of Lululemon, a Canadian brand that records the highest sales in the global leggings market, were created. The leggings measurements are presented in

Table 1. The size of the leggings pattern used in this study.

Measurement Area	Inseam	Waist Circumference	Hip Circumference	Thigh Circumference	Knee Circumference	Calf Circumference	Ankle Circumference	Total Length
Size (cm)	65	60	85	45	30	25	18	82

. The body measurements of the avatar wearing the 3D virtual try-on garment were based on the International Organization for Standardization (ISO) standard size ISO 8559-1 [29] for Small (S), with a bust circumference of 84 cm, waist circumference of 64 cm, hip circumference of 90 cm, and a global average height of 163 cm. This standardized avatar is widely adopted in virtual simulation studies due to its representativeness and compatibility with existing sizing systems. However, it does not reflect the diversity of real-world body types (e.g., individuals with obesity or shorter/taller stature), which represents a limitation of this study. To improve the generalizability and applicability of future results, expanded size ranges and more diverse body shapes should be considered in subsequent research.

The 3D virtual try-on garment was completed through avatar creation, pattern input, sewing, and layout, followed by a final simulation. During the virtual try-on, the relevant values were entered into the Fabric Property window of the CLO 3D program to implement the physical properties of the fabric. The fabric of Lululemon leggings was designed with varying spandex contents to accommodate body movements and comfort during sports activities. The material with the highest spandex content from Lululemon included 80% nylon and 20% spandex, whereas the material with the lowest spandex content included 87% nylon and 13% spandex. These two types of fabrics were used in this study. Fabric A has a higher spandex content and thus high stretchability, allowing it to closely fit the body such that the clothing pressure applied to the inside of the wearer’s body may be higher when the wearer remains stationary or performs small movements in daily activities. However, when the wearer performs large movements, the material stretches well along the body, allowing the wearer to feel more comfortable. Fabric B has a relatively lower spandex content, resulting in lower stretchability than Fabric A, and it may feel slightly looser or softer than Fabric A when the wearer remains still. However, when the wearer performs larger movements, there is less stretching in Fabric B than Fabric A, which may cause the wearer to experience more pressure on the outside of the body. However, this pressure may be preferred by some wearers because it provides a more stable fit during exercise movements.

Five beginner-friendly yoga poses were selected, including the Mountain, Cobra, Lunge pose, Plow, and Forward Fold poses. Figure 1 summarizes the study procedure, and Figure 2 presents the five yoga poses used. These poses were selected to include various joint flexion areas and body movement ranges. Prior biomechanical studies have demonstrated that such poses engage distinct muscle groups and involve varying degrees of joint range of motion, particularly in the hip, knee, and ankle joints (e.g., Liu et al., 2021 [30]; Pinto et al., 2022 [31]). Accordingly, they were deemed suitable for assessing posture-dependent variations in clothing pressure. When performing yoga poses while wearing leggings, the body parts used and the range of their movements vary depending on the type of movement, which affects the magnitude of clothing pressure. At body parts where the leggings fabric stretches during movement, the clothing pressure decreases, whereas at body parts where the leggings contract or are pulled, the clothing pressure increases. Moreover, regardless of the type of material (or stretchability), clothing pressure may increase at the body parts that make contact with the floor during floor-based movements. Figure 1 illustrates the overall workflow of this study. After pattern making, fabric composition, virtual try-on, and simulation of five yoga poses, digital clothing pressure was calculated. In addition, a data input and management step was included, in which material properties (fiber composition, blend ratio, knit structure) were entered into CLO 3D, and the resulting pressure data were exported in CSV format for further analysis. This reflects the proposed optimization functions related to data handling (Section 4.3, point 5).

2.2. Clothing Pressure Measurement

Clothing pressure refers to the force exerted by garments worn on the body from various angles, which varies significantly depending on the stretchability of the fabric and the wearer’s movements. Accordingly, this study measured clothing pressures for two types of fabric with different stretchabilities and five yoga poses. Fabric A, which has a higher spandex content, is used for leggings recommended for activities in which flexibility and stretchability are important, such as yoga, Pilates, and stretching. Fabric B, which has a relatively lower spandex content, is used for leggings that require a fit to provide body support during high-intensity exercises, including weight training and CrossFit.

Six body parts were selected for measuring the clothing pressure in CLO 3D: the waist, hip, thigh, knee, calf, and ankle (Figure 3). The waist (P1) was measured around the navel where the torso curves inward. The waist is one of the most critical body parts for fitting leggings, as the degree of pressure applied to the waist significantly affects the wearer’s overall comfort; thus, it should be neither too tight nor too loose. The hip (P2) is important for evaluating the support provided by the leggings to the wearer’s hips and was measured at the widest part of the hip. The thigh (P3) experiences many significant movements during exercise and was measured at the mid-thigh. The knee (P4) serves as the central joint for lower-body movements and requires free flexion and extension. Therefore, it was measured in front of the knee in the flexed position. The calf (P5) and ankle (P6) require comfort during movement as well as adequate coverage without being too tight or too loose.; they were measured at the widest part of the calf and just above the ankle, respectively.

The virtual try-on clothing pressure was analyzed using the stress distribution feature provided in the 3D view of CLO 3D, and the clothing pressure values for each of the six body parts were obtained by clicking on the respective areas with the mouse. Because the clothing pressures in this study were measured in a virtual environment, clothing pressure kPa(kilopascal) refers to the virtual kPa in the following sections [32]. In CLO 3D, clothing pressure levels are visualized in the form of a color map. Blue (low pressure) represents minimal contact between the garment and the body, green–yellow (medium pressure) represents an appropriate level of wearability and a relaxed fit, and red (high pressure) represents strong compression that may cause discomfort. To enhance the consistency of data collection, each measurement was performed three times per pose, and the average value was used for analysis. Additionally, visual inspection of the stress map was conducted to support the interpretation of the numerical results. While this method offers a practical approach for evaluating pressure distribution in virtual garments, it should be noted that CLO 3D currently lacks calibration features and third-party validation tools to verify the accuracy of its pressure estimates.

2.3. Data Analysis

To assess how well the CLO 3D virtual fitting system represents garment fit, the clothing pressures at the six body parts of the leggings were qualitatively and quantitatively analyzed for the five yoga poses (see Figure 2). A qualitative analysis was conducted based on the stress distribution visualized in colors provided by the 3D view function of the software. The stress color map was analyzed from all directions (front, side, and back) of the virtual try-on sample, and the differences in stress colors for each body part were observed according to the spandex content of the fabric while performing the five yoga poses. For quantitative analysis, the clothing pressure values measured for each body part in the software were numerically compared according to yoga pose and fabric type. Based on these qualitative and quantitative analyses, an optimized material and fit (clothing pressure) simulation function of the 3D fashion design software was proposed.

3. Results

3.1. Qualitative Analysis of Clothing Pressure Differences According to Yoga Poses and Spandex Content

Table 2. Garment Fit Maps for five yoga poses and two types of fabrics (A: Nylon 80% and Spandex 20%, B: Nylon 87% and Spandex 13%).

Position	Mountain Pose		Cobra Pose		Lunge Pose		Plow Pose		Forward Fold
	A	B	A	B	A	B	A	B	A	B
Front
Side
Back

Note. Color scale for clothing pressure visualization: Blue = Low pressure, Green–Yellow = Medium pressure, Red = High pressure.

presents the stress color maps of the five yoga poses provided by the CLO 3D view function. The different colors on the map allow for the identification of clothing pressure levels by body part based on the wearer’s movements. Regardless of the type of yoga pose, medium clothing pressure was generally observed in the waist, thigh, and calf areas. The hip area tended to show relatively high clothing pressure, while the ankle and knee areas tended to exhibit relatively low clothing pressure. The Composition Customization function of CLO 3D allows the selection of the blend ratio of fashion materials in percentages (%). However, despite the difference in spandex content between the two fabrics used in this study, it was difficult to qualitatively observe the differences in clothing pressure through images A and B presented in Table 2. In addition to the integrated comparison shown in Table 2, Figure 3, Figure 4, Figure 5, Figure 6 and Figure 7 separately illustrate the simulated clothing pressure maps for each yoga pose. These figures allow for a clearer visualization of posture-specific stress distributions.

Brubacher et al. (2023) [33] demonstrated that clothing pressure values measured by virtual fitting software do not always align with actual physical measurements, especially in the case of compression garments. In line with their findings, this study also found that CLO 3D failed to reflect mechanical differences in fabric properties, as the stress color maps showed no distinguishable variation between materials with different spandex content. These results support the notion that the current CLO 3D system has limitations in representing fabric-specific tension across varying body postures.

3.2. Quantitative Analysis of Clothing Pressure Differences According to Yoga Poses and Spandex Content

According to the above qualitative analysis, there is no significant difference in clothing pressure according to the type of material, that is, spandex content. This section presents a quantitative analysis based on clothing pressure values obtained by clicking on the same 3D virtual fitting system.

Table 3. Digital clothing pressure data collected by virtual try-on.

Position	Digital Clothing Pressures (Unit: kPa)
	Mountain Pose		Cobra Pose		Lunge Pose		Plow Pose		Forward Fold
Fabric a	A	B	A	B	A	B	A	B	A	B
P1 (Waist)	36.02	36.02	36.11	36.11	35.48	35.48	36.18	36.18	37.55	37.55
P2 (Hip)	63.83	63.83	66.03	66.03	65.96	65.96	68.69	68.69	70.28	70.28
P3 (Thigh)	38.53	38.53	39.15	39.15	68.31	68.31	36.98	36.98	40.03	40.03
P4 (Knee)	37.97	37.97	38.38	38.38	38.64	38.64	33.52	33.52	33.20	33.20
P5 (Calf)	43.93	43.93	44.81	44.81	45.30	45.30	45.04	45.04	45.61	45.61
P6 (Ankle)	25.51	25.51	20.95	20.95	23.33	23.33	20.96	20.96	25.81	25.81

^a Fabric A: Nylon 80% and spandex 20%; Fabric B: Nylon 87% and spandex 13%.

presents the clothing pressure values for Fabrics A and B for each body part during the five yoga poses. The software could not distinguish the differences in clothing pressure between the two fabrics for all body parts across all five yoga poses. This outcome suggests a potential limitation of the CLO 3D simulation engine, which may lack sufficient sensitivity to detect mechanical differences arising from subtle variations in spandex content (e.g., 20% vs. 13%). Although, in theory, differences in stretchability should result in corresponding variations in clothing pressure, such distinctions were not observed in the virtual simulation. This finding highlights the need for further enhancement of the software’s ability to simulate material-specific behavior with greater precision. Given that the pressure values were identical across all conditions, no statistical tests (e.g., t-test or ANOVA) were conducted. Instead, the complete lack of variation itself was interpreted as evidence of the simulation engine’s insufficient responsiveness to subtle differences in fabric properties.

Figure 4. Representative CLO 3D simulation images for five yoga poses. (a) Mountain Pose, (b) Cobra Pose, (c) Lunge Pose, (d) Plow Pose, (e) Forward Fold. Color scale: Blue = Low pressure, Green–Yellow = Medium pressure, Red = High pressure.

Figure 4. Representative CLO 3D simulation images for five yoga poses. (a) Mountain Pose, (b) Cobra Pose, (c) Lunge Pose, (d) Plow Pose, (e) Forward Fold. Color scale: Blue = Low pressure, Green–Yellow = Medium pressure, Red = High pressure.

The clothing pressure for each body part was numerically compared according to the yoga poses, and the results are presented in Figure 5. Regardless of the type of movement, the clothing pressure was the highest in the following order: hip (mean, 66.96; standard deviation [SD], 2.27), calf (mean, 44.94; SD, 0.64), thigh (mean, 49.96; SD, 14.07), knee (mean, 36.34; SD, 2.45), waist (mean, 36.27; SD, 0.69), and ankle (mean, 23.31; SD, 2.35). These results could not be obtained from the previous qualitative analysis.

Figure 5. Numerical comparison of clothing pressure by body part according to yoga poses, expressed in kPa.

Figure 5. Numerical comparison of clothing pressure by body part according to yoga poses, expressed in kPa.

The analysis of the specific yoga poses revealed the following results. First, the Mountain Pose, which is a basic alignment pose with the body fully extended, exhibited a relatively even pressure distribution. The lowest pressure (25.51 kPa) was observed at the ankle, whereas the highest pressure (63.83 kPa) was observed at the hip. Although, in this pose, the highest clothing pressure occurred at the hip among the six body parts, the pressure on the hip was lower than that of the other four poses because this pose did not involve significant stretching of the body. For the remaining body parts, such as the waist, thigh, knee, and calf, clothing pressure ranged from 36.02 kPa (waist) to 43.93 kPa (calf), with similar values.

Second, the Cobra Pose involves lifting the upper body while lying on the stomach, stretching the abdominal area, and pressing the front of the lower body against the floor. The highest clothing pressure (66.03 kPa) occurred at the hip, which stretches as the upper body is lifted, while a relatively lower clothing pressure (36.11 kPa), similar to the level seen in the Mountain Pose, occurred at the waist, which stretches less above the hip. In addition, because of the overall leg stretch during the pose, higher clothing pressures were observed at the thigh (39.15 kPa), knee (38.38 kPa), and calf (44.81 kPa) compared with the Mountain Pose. However, because less force is applied to the ankle during the Cobra Pose, the clothing pressure at the ankle is lower in the Cobra Pose than in the Mountain Pose (20.95 kPa).

Third, the Lunge pose is a pose where one leg is deeply bent forward, and lower body stretching and muscle strength are critical. Because the movements of the front and back legs differ during lunging, the clothing pressure applied to each part also varies. Compared to the Cobra Pose, where the hip is greatly flexed, the clothing pressure at the hip is lower in the Lunge Poses (65.96 kPa) and the pressure at the waist is slightly decreased (35.48 kPa). Moreover, the back thigh, which is stretched out, experienced the highest pressure among the five poses (68.31 kPa). High levels of clothing pressure were also observed in the knee (38.64 kPa), calf (45.30 kPa), and ankle (23.33 kPa) of the extended back leg along with the thigh.

Fourth, the Plow Pose involves lying flat on the back and bringing the legs over the head, causing significant bending from the upper body to the waist. The highest clothing pressure (68.69 kPa) occurred at the hip, which serves as the central point where the body’s upper and lower parts reverse positions. A higher clothing pressure was also observed at the waist (36.18 kPa) compared with all other poses, except the Forward Fold. In comparison to the upper parts of the lower body, the lower parts exert less force during the Plow Pose, which leads to generally lower clothing pressures compared to the other poses (thigh: 36.98 kPa; knee: 33.52 kPa; calf: 45.04 kPa; and ankle: 20.96 kPa).

Fifth, the Forward Fold involves standing with the legs straight and bending only the upper body downward, causing the waist to fold and the hip to stretch significantly. As a result, during the Forward Fold, clothing pressures were the highest at the waist and hip among all five yoga poses (37.55 kPa and 70.28 kPa, respectively). As the upper body is bent and the leggings stretch significantly near the hip, a high clothing pressure of 40.03 kPa was observed at the thigh. Since the knee remains straight, the basic clothing pressure is maintained at a relatively low level of 33.20 kPa, while relatively high clothing pressures were observed at the calf and ankle supporting the body weight (45.61 kPa and 25.81 kPa, respectively).

Meanwhile, as a variable that can predict the overall comfort of the yoga clothing wearer, the total clothing pressure was calculated for the five yoga poses (by summing the clothing pressures for all body parts). The results are presented in Figure 6. The total clothing pressure for Lunge Pose was the highest at 277.02 kPa, followed by Forward Fold at 252.48 kPa, while Mountain Pose and Cobra Pose had similar levels of 245.79 kPa and 245.43 kPa, respectively. The Plow Pose had the lowest total clothing pressure at 241.37 kPa. Kweon (2012) [34], who studied the subjectively perceived clothing pressure of tight pants based on the wearer’s movements, has found that wearers experienced the greatest pressure in poses, such as the Lunge Pose, where the legs are bent, and relatively less pressure in poses, such as the Mountain Pose, where the body remains upright, thereby supporting the findings of this study.

Figure 6. Total clothing pressure for each yoga pose, expressed in kPa (mean ± SD, with error bars).

Figure 6. Total clothing pressure for each yoga pose, expressed in kPa (mean ± SD, with error bars).

When implementing moving poses in the CLO 3D virtual simulation system, the friction between the fabric and the body or between fabrics, as well as the direction in which the fabric stretches, may differ from reality. This study confirmed that the current system simplifies the way fabric adheres to the body and changes with the wearer’s movements in compression garments or tight-fitting clothing, such as leggings. Therefore, to help users of 3D fashion design software more accurately understand the fit as well as the overall appearance and tactile properties of the actual garment and predict the wearer’s comfort, there is a need for software optimization that considers material. Since the CLO 3D system failed to distinguish any pressure variation based on fabric differences, conducting physical wear trials at this stage would not yield meaningful validation outcomes. Future studies should include such tests after addressing the current limitations of CLO 3D in detecting material-dependent pressure changes.

4. Discussion

4.1. Limitations of CLO 3D in Simulating Material Differences

The present study demonstrated that the CLO 3D simulation system failed to reflect differences in clothing pressure according to spandex content. Despite the theoretical expectation that higher spandex content should result in greater stretchability and altered compression, the virtual fitting engine produced identical pressure values for both fabrics. This outcome indicates insufficient sensitivity of the current simulation engine to fabric-specific mechanical behavior. Similar findings were reported by Brubacher et al. (2023) [33], who highlighted discrepancies between virtual fit technology and the actual mechanical performance of compression garments, further confirming that the CLO 3D platform currently lacks accuracy in replicating fine material-dependent variations.

4.2. Posture-Dependent Pressure Distribution and Comparison with Previous Studies

Although fabric differences were not captured, posture-related variations in clothing pressure were clearly observed across yoga poses. For example, hip regions consistently showed the highest pressure, while the ankle exhibited the lowest values. These patterns are consistent with the results of Kweon (2012) [34], who found that postures involving deep flexion, such as Lunge pose, induced greater perceived pressure compared to upright postures. Thus, CLO 3D could capture the general effects of body posture on pressure distribution, but it was unable to distinguish subtle differences in material properties. This highlights the system’s partial utility—reliable for identifying gross posture-dependent trends but limited in evaluating fabric-specific responses.

4.3. Suggestions for Optimized Simulation Functions in CLO 3D

CLO 3D is a useful tool in fashion design and simulation; However, it has limitations in realistically evaluating garments because of insufficient data on the complex properties of materials and wearer comfort. To overcome this problem, it is necessary to develop a system that integrates sophisticated physics engines with environmental data. To accurately measure the clothing pressure of spandex and nylon blended fabrics in CLO 3D, it may be useful to add the following functions.

First, although CLO 3D currently has the function of setting the blend ratio of spandex to nylon, it should be further developed such that various physical properties can be calculated based on this blend ratio. Because stretchability and recovery are important elements of spandex and similar materials developed thereafter, the system should be capable of simulating clothing pressure and evaluating the permanent deformation resulting from repetitive movements of the wearer. As shown in Figure 7, this study suggests expanding the material data input function in CLO 3D to provide users with detailed input options that allow them to include the unique properties of special materials. Users would be able to directly enter or select properties from a library of options, including fabric knitting methods, elastic modulus, stretch ratio, and recovery. Adding this function would enable an accurate clothing pressure analysis of spandex and nylon blend fabrics and significantly improve the efficiency of designing sportswear, medical compression garments, and custom fashion apparel, as well as the quality of the final products.

Figure 7. (a) Current CLO 3D (2024.2.214 version); (b) Proposed new option 1: Entering material physical properties.

Figure 7. (a) Current CLO 3D (2024.2.214 version); (b) Proposed new option 1: Entering material physical properties.

Second, in garments with multiple layers of fabric, interactions occur between the materials, for example, friction between the lining and outer fabric and the movement of padding. However, the current CLO 3D can only simulate single-layer materials; therefore, its accuracy is limited when simulating mixed fabrics or complex garments. This is particularly challenging for simulating functional materials such as those with water-repellent, moisture-wicking, windproof, and thermal properties, which are difficult to represent in the current CLO 3D system. To enable the customized simulation of special materials in CLO 3D, the system should be designed to precisely reflect the unique physical properties and special functions of these materials. Thus, the second proposed function, shown in Figure 8, provides a database in collaboration with fabric manufacturers that can be updated in real time in the Fabric menu of the Library, thereby enhancing the physical property information of technical textiles. This involves improving the physical engine to digitally simulate the fine movements and properties of fibers and adding a customized simulation model for technical textiles.

Third, the current CLO 3D system cannot simulate the thermal properties of clothing/materials (such as temperature regulation, breathability, and sweat absorption) as indicators of comfort under various climatic conditions. Thus, it is difficult to predict the thermal comfort of garments in hot and cold weather. This thermal physiological comfort is affected by the thermal properties of the clothing/materials mentioned earlier, which the current CLO 3D fails to consider, and by external environments such as external temperature and humidity, the wearer’s activity level, and body constitution, as well as the climate in the microspace between the skin and garment. Accordingly, the third proposed function, shown in Figure 9, allows users to set the simulation environment (indoor, outdoor, seasonal, etc.) as the most fundamental factor and presents the thermal comfort of the garment in that environment. This function would offer a wide range of adjustable temperature and humidity settings to match the environmental settings of the human-specific artificial environmental chamber system. It would then provide a simulation of the wearer’s body temperature distribution and additional data, such as heat flow and sweat absorption rate, integrating garment-specific data with wearer-centered data. For example, this would inform users that clothing made from materials with hygroscopicity and breathability below a certain level in a summer outdoor environment (temperature 35 °C, humidity 70%) would have low comfort. This function would allow for the simulation of environmental sensitivity in special materials (e.g., phase change material: materials that change phase depending on the environment; chromic materials: materials that change color depending on the environment). It would also enable the prediction of issues such as color fading and physical property deterioration owing to repeated wear in different environments, even without actual UV or abrasion testing, making this function particularly useful for the design of sportswear or outdoor clothing.

Fourth, a machine learning model should be introduced to predict user comfort based on actual wear experience data collected from various users (Figure 10). By leveraging machine learning, the expected performance of fabrics—such as durability, stretchability, and friction—can be rapidly estimated from the material properties entered by the user before running the simulation. To implement this function, the system would collect user-specific data including body shape, activity type (e.g., yoga, running), and environmental conditions (e.g., temperature, humidity). A supervised learning algorithm could then be trained to associate these variables with historical material performance data and subjective comfort feedback. For instance, a user engaging in high-flexion activities in warm, humid environments might be recommended a fabric with high recovery and breathability, while someone exercising in colder climates may be guided toward materials with superior thermal insulation. This function could form the basis of a personalized material recommendation system that suggests fabrics or garment designs most suited to a user’s purpose and environment.

Fifth, an additional function proposed in this study is to provide data in a table format or add a data export function (e.g., Excel and CSV) in CLO 3D to enable efficient and quick analysis for users (Figure 11). Data collection and management can be facilitated by supporting the upload of material profiles based on actual experimental data, allowing the input of numerical data in CSV or JSON formats, and integrating experimental data with virtual data. Furthermore, a dedicated dashboard can be implemented to visualize the material input and resulting data more effectively through UI/UX enhancements. If simulation results are presented as graphs or 3D animations, users would be able to generate and analyze data solely within CLO 3D, eliminating the need for external software such as Excel.

4.4. Limitations and Future Work

This study has several limitations. The analysis was restricted to a small-size female avatar and two types of nylon–spandex fabrics, which may limit the generalizability of the findings. Furthermore, subjective comfort evaluations were not included, making it difficult to directly link virtual pressure data with wearer perception. Future research should expand the range of body types, sizes, and materials analyzed, and incorporate empirical wear trials to validate simulation results. Combining subjective evaluations with CLO 3D data will provide a more comprehensive understanding of the relationship between simulated clothing pressure and actual wear comfort.

5. Conclusions

This study evaluated the capability of CLO 3D virtual fitting software to represent clothing pressure for compression wear made from nylon–spandex blends across five yoga poses. The results demonstrated that CLO 3D failed to differentiate pressure variations according to spandex content, indicating limitations in its sensitivity to material-specific mechanical properties. Nevertheless, posture-related pressure differences were evident, with the hip consistently exhibiting the highest pressure and the ankle the lowest. Among the poses, the Lunge pose produced the greatest total clothing pressure (277.02 kPa), while the Plow pose resulted in the lowest (241.37 kPa).

To address these shortcomings, five optimization functions were proposed: (i) material property input, (ii) technical textile databases, (iii) environmental condition settings, (iv) AI-based comfort prediction, and (v) data management tools. These improvements would enhance the scientific validity and practical applicability of CLO 3D, enabling more accurate simulations for functional apparel design, including sportswear and medical compression garments.

While this study provides foundational insights, its scope was limited to a small-size female avatar and two fabric types, without incorporating subjective comfort evaluations. Future work should expand to diverse body types, fabric compositions, and empirical wear trials to validate virtual results. Overall, this research highlights both the potential and the limitations of CLO 3D and provides a pathway toward developing a more robust simulation tool for functional textile and apparel innovation. Accurate simulation is not only a technical matter but also a critical foundation for broader applications. In sportswear, reliable pressure prediction can guide the design of garments that enhance athletic performance and reduce injury risk. In medical contexts, accurate virtual evaluation of compression levels is directly linked to patient safety and treatment efficacy in products such as medical stockings and rehabilitation wear. Furthermore, in the apparel industry, more precise simulation contributes to sustainable product development by reducing the number of physical prototypes, minimizing material waste, and shortening lead times. Therefore, the proposed improvements in CLO 3D are expected to generate a wider impact that extends beyond virtual prototyping, influencing health, sustainability, and innovation in fashion technology.