iStyleAR - Virtual Try-On Using Augmented Reality

Abstract

In the age of digital transformation and online purchasing, customers are normally susceptible to difficulty in imaginingthewayclotheswouldappearandfitpriortopurchase. The present paper describes iStyleAR - Virtual Try-On Using Augmented Reality, the web-based Augmented Reality (AR) try- on, which allows users to virtually try clothes in real time with only a webcam. The system integrates Mediapipe Pose Detection, OpenCV, and a Flask server to properly superimpose apparel images on the body of the user. The frontend is a responsive React application, based on Firebase authentication, that offers aninteractiveinterface,i.e.outfitswitchingbasedongesture sand avirtualwardrobepersonalizedtotheuser.Asaprototypethatis lightweight and can be accessed on the browser, iStyleAR shows howARcanmakeonlineshoppingmoreconvenientandpersonal styling possible without any specialized hardware. The future extensions are 3D garment visualization, smart size suggestions and enhanced e-commerce inclinations to enable the all-inclusive fashion experience.

Introduction

Augmented Reality (AR) is rapidly transforming the retail and fashion industries by overlaying virtual elements onto real-world environments, improving customer engagement and helping users visualize products before purchase. Traditional shopping—whether online or offline—often fails to provide realistic garment previewing, resulting in uncertainty, dissatisfaction, and high product return rates. Motivated by these challenges, the proposed AR Clothes Try-On system aims to deliver an accessible, real-time virtual fitting solution using computer vision and gesture-based interaction.

The system uses Python, OpenCV, MediaPipe, and CVZone to detect human body landmarks and dynamically overlay clothing images onto the user in live video. Users can switch between outfits using simple hand gestures, eliminating the need for keyboards or controllers. Unlike high-end AR solutions dependent on 3D modeling or specialized equipment, this system demonstrates that practical AR try-on experiences can be achieved using only a webcam and standard computing hardware.

A review of prior literature shows that AR significantly boosts customer engagement, purchase intent, and conversion rates in fashion retail. Existing systems often face limitations such as dependency on complex 3D simulations, high computational cost, or limited gesture interaction. The proposed model addresses these gaps by combining accurate pose estimation with lightweight gesture recognition, creating a more accessible and interactive AR try-on experience suitable for personal styling and e-commerce integration.

The methodology includes real-time video capture, landmark detection, dynamic garment scaling and placement, and gesture-controlled apparel selection. A web-integrated version using React, Flask, and Firebase enables cloud storage, user authentication, and online wardrobe features. Experimental results show stable garment alignment, responsive gesture control, and smooth real-time performance. User feedback indicates high satisfaction with system speed, realism, and hands-free interaction, though low-light environments can affect accuracy.

Overall, the system proves that affordable AR-based virtual try-on solutions can enhance online shopping, reduce returns, and improve user confidence. Future developments may include 3D garment simulation, personalized avatars, real-time body measurements, and deeper integration with e-commerce platforms.

Conclusion

The paper described an Augmented Reality (AR) Clothes Try-On System created with the help of Python, OpenCV,and CVZone to create an interactive and lifelike virtual try-on experience. The system is successful in integrating pose de- tection and gesture recognition into dynamic clothing overlay and hands-free navigation between clothes. The system shows that it is possible to make AR-based fashion visualizationboth efficient and accessible without the use of high-end or specialized equipment, as it has stable performance and canbe rendered smoothly. Virtualtry-ontechnologydrivenbyARcanbecomea great way to shop online because it can greatly increase user interaction,purchaseconfidence,andlowertherateofproduct returns.Thegapbetweenonlinebrowsingandin-storetrialcan be filled by the proposed system to emphasize how AR could revolutionize fashion retail ecosystem using personalized and interactive experience. On balance, this work provides the basis of the future research and development of the augmented reality applica- tionsinthefashionretail.Itwillshowhowtheintersection of computer vision and AR can develop meaningful user- centered innovations to reinvent digital shopping experiences and personal styling.

References

[1] Azuma,R.T.(1997).Asurveyofaugmentedreal- ity. Presence: Teleoperators & Virtual Environments, 6(4), 355–385. [2] Billinghurst,M.,Clark,A.,Lee,G.(2015).Asur- vey of augmented reality. Foundations and Trends in Hu- man–Computer Interaction, 8(2-3), 73–272. [3] Sharma,N.,Singh,A.,Singh,M.,Walia,A. (2025). Augmented reality-based virtual try-on for clothing:A real-time user-centric approach. International Journal of Novel Research and Development (IJNRD), 10(5), Article IJNRD2505006. [4] Cao,Z.,Simon,T.,Wei,S.-E., Sheikh,Y.(2017). Realtimemulti-person2Dposeestimationusingpartaffinity fields. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 7291–7299). IEEE. [5] Batool, R.,Mou, J. (2023). A systematic literature reviewandanalysisoftry-ontechnology:Virtualfittingrooms. Heliyon, 9(9), e21662. Elsevier. [6] Chen, C., Ni, J.,Zhang, P. (2024). Virtual try-on systemsinfashionconsumption:Asystematicreview.Applied Sciences, 14(24), 11839. [7] Osokin, D. (2018). Real-time 2D multi-person pose estimation on CPU: Lightweight OpenPose. arXiv preprint arXiv:1811.12004. [8] Wu,Y.,Liu,H.,Lu,P.,Zhang,L.,Yuan,F.(2022). Design and implementation of virtual fitting system based on gesture recognition and clothing transfer algorithm. Scientific Reports, 12(1), 17890. [9] Gavgiotaki, D., Ntoa, S., Margetis, G., Apostolakis, K. C.,Stephanidis, C. (2023). Gesture-based interaction for AR systems: A short review. In Proceedings of the 15th ACM International Conference on PErvasive Technologies Related toAssistiveEnvironments(PETRA’23)(pp.377–384).ACM. [10] Sarkis,N.,JabbourAlMaalouf,N.,Saliba,E.,Azizi,J. (2025).Theimpactofaugmentedrealitywithinthefashionin- dustryonpurchasedecisions,customerengagement,andbrand loyalty. International Journal of Fashion Design, Technology and Education. Advance online publication. [11] Kavyashree,L.,Sharma,K.,Raj,R.,Sarapali,M. C.,Chandana, P. M. (2025). A review on virtual try-on systems using augmented reality and artificial intelligence in fashion retail. International Journal of Science, Engineering and Technology (IJSET), 13(3), 1–6. [12] Kannaiah, D.,Shanthi, R. (2015). The impact of augmented reality on e-commerce. Journal of Marketing and Consumer Research, 13, 55–62.

Copyright

Copyright © 2025 Kush Dastane, Pranav Chavan, Pushpak Khade, Sudhanshu Ray. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.