Virtual Try-ON Clothes

Abstract

The increasing popularity of e-commerce and the desire for personalized shopping experiences have led to the development of virtual trial room applications. In this research paper, we present a novel approach to outfit prediction in a web-based virtual trial room. Our system utilizes machine learning techniques to analyze input images of shirts and pants and provides accurate outfit recommendations to users. The application, developed using Python, Flask, HTML, CSS, JavaScript, Bootstrap, and OpenCV, offers users the ability to virtually try on different outfits without physically trying them on. We discuss the design, implementation, and evaluation of our system, highlighting its effectiveness in accurately predicting outfit combinations and improving the user experience. Digital try-on systems for e-commerce have the potential to change people’s lives and provide notable economic benefits. However, their development is limited by practical constraints, such as accurate sizing of the body and realism of demonstrations. We enumerate three open challenges remaining for a complete and easy-to-use try-on system that recent advances in machine learning make increasingly tractable. For each, we describe the problem, introduce state-of-the-art approaches, and provide future directions.

Introduction

I. PROBLEM STATEMENT

E-commerce has grown at a rapid pace in recent years. Consumers today are more likely to shop online than to visit a retail store. The situation is much more complicated, however, when it comes to buying clothes. People need to know how a garment fits on them, how it looks, and how it feels.

II. INTRODUCTION

The emergence of virtual trial room applications has revolutionized the way people shop for clothes online. By enabling users to virtually try on different outfits, these applications bridge the gap between the online and offline shopping experience. In this paper, we present a machine learning-based approach to outfit prediction in a web-based virtual trial room, enhancing the accuracy and convenience of outfit selection for users.

Digital try-on systems for e- commerce have the potential to change people’s lives and provide notable economic benefits. However, their development is limited by practical constraints, such as accurate sizing of the body and realism of demonstrations. We enumerate three open challenges remaining for a complete and easy-to-use try-on system that recent advances in machine learning make increasingly tractable. For each, we describe the problem, introduce state-of-the-art approaches, and provide future directions.

VI. ALGORITHM AND LIBRARY USED

A. Haar Cascade Classifier Algorithm

1. A machine Learning Algorithm used for object detection, particularly for detecting faces
2. The Haar Cascade Classifier algorithm works by using a set of Haar-like features & a cascade of waek claasifier trained with the ADAboost Algorithm.
3. It Evaluates these features at multiple sacles & positions within a sliding window framework , efficiently rejecting non face regions early on.
4. The algorithm strength lies in the ability to quickly detect faces in real time application.

Haar cascade is an algorithm that can detect objects in images, irrespective of their scale in image and location.

This algorithm is not so complex and can run in real-time. We can train a haar-cascade detector to detect various objects like cars, bikes, buildings, fruits, etc.

B. OpenCV Library

OpenCV is a Python library that allows you to perform image processing and computer vision tasks. It provides a wide range of features, including object detection, face recognition, and tracking.

OpenCV is an open-source software library for computer vision and machine learning. The OpenCV full form is Open Source Computer Vision Library. It was created to provide a shared infrastructure for applications for computer vision and to speed up the use of machine perception in consumer products. OpenCV, as a BSD- licensed software, makes it simple for companies to use and change the code. There are some predefined packages and libraries that make our life simple and OpenCV is one of them.

VII. IMPLEMENTATION DETAILS

We provide detailed insights into the implementation of our virtual trial room system. This includes the front-end design and user interface considerations, back-end functionality integration, and the integration of the machine learning model within the application. We discuss the challenges encountered during implementation and the strategies employed to overcome them.

The Webpages for the project is developed with the frontend technologies like html an CSS which are used to create the user friendly interface and handle client site interactions. To give the extra look and feel to the website we use the bootstrap framework. Bootstrap library is use for responsive design and predefined styles. For the backend we use Python. We use an open source flask framework which is written in python. We use computer vision library which is open cv library use for image and video processing

A. Evaluation

We present the evaluation methodology used to assess the performance of our system. We discuss the dataset used for training and testing, the evaluation metrics employed, and the experimental results obtained. The evaluation demonstrates the accuracy of outfit predictions and the overall effectiveness of the system.

B. Input

• Clothes Image
• Target body

VIII. RESULTS AND DISCUSSION

We present and analyze the results obtained from the evaluation, emphasizing the performance and usability of the virtual trial room system. We discuss the significance of accurate outfit predictions in improving the user experience and the potential impact on the e-commerce industry.

IX. FUTURE SCOPE

A. Adding more clothes option.

B. Integrating color customization.

C. Improving the accuracy of clothing alignment.

D. Encourage the audience to think about how this technology can be further develop and applied in different domains.

Conclusion

In this work we have been focusing on customer interface of the virtual dressing room. We will developing a 3-D cloth scanner which would be interesting to test in usability and user experience study. We summarize the key contributions of our research and discuss future directions for enhancing virtual trial room systems. We highlight the potential for incorporating additional features, such as personalized recommendations and augmented reality, to further improve the user experience. In this work we have been focusing on customer interface of the virtual dressing room. We will developing a 3-D cloth scanner which would be interesting to test in usability and user experience study.

References

[1] Chao Lin, Zhao Li, Sheng Zhou, Shichang Hu, Jialun Zhan Linhao Luo, Yuan He (2022) RMGN: A Regional Mask Guided Network for Parser-free Virtual Try-on [2] Cheng JCP, Chen K, Chen W (2017) Comparison of Marker-Based and Markerless AR: A Case Study of An Indoor Decoration System. In: Lean and Computing in Construction Congress - Volume 1: Proceedings of the Joint Conference on Computing in Construction. Heriot-Watt University, Heraklion, Crete, Greece, pp 483– 490 [3] Ge Y, Song Y, Zhang R, Ge C, Liu W, Luo P (2021) Parser-Free Virtual Try-on via Distilling Appearance Flows. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Nashville, TN, USA, pp 8481–8489 [4] Groubert J, Langlotz T, Zollmann S, Regenbrecht H (2017) Towards Pervasive Augmented Reality: Context-Awareness in Augmented Reality. IEEE Trans Visual Comput Graphics 23(6):1706–1724. https://doi.org/10.1109/TVCG.2016.254372

Copyright

Copyright © 2023 Prof. Pooja Parikh, Reshma Kale, Vaishnavi Bidarkar, Vaishanavi Sarode, Soni Singh. 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.