Virtual Try on Application and Parameter Analysis Based on Clothing Warping and Masking Technology

Chunbo Feng

2024

Abstract

Virtual Fitting represents a pivotal application of technology, facilitating seamless online clothing purchases by enabling users to visualize garments on their bodies. As a focal point within the image generation domain, it holds immense commercial value, garnering substantial attention. This paper delves into the examination of parameters and training sessions' impact on Diffusion-based Conditional Inpainting for Virtual Try-ON (DCI-VITON), elucidating the efficacy of individual blocks and the overall model. Leveraging garment warping and masking techniques, the method preserves coarse character image structures while harnessing the potent generative capabilities of diffusion models to refine results. Experimental findings highlight the nuanced effects of various parameters on model performance, showcasing its superiority over alternative virtual fitting methods, particularly in capturing clothing details and styles. The model's ability to produce realistic, intricately detailed morphed garments with consistent visual quality underscores its significance. This project serves as a crucial reference for future virtual fitting techniques grounded in diffusion models.

Paper Citation

in Harvard Style

Feng C. (2024). Virtual Try on Application and Parameter Analysis Based on Clothing Warping and Masking Technology. In Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI; ISBN 978-989-758-713-9, SciTePress, pages 328-332. DOI: 10.5220/0012937600004508

in Bibtex Style

@conference{emiti24,
author={Chunbo Feng},
title={Virtual Try on Application and Parameter Analysis Based on Clothing Warping and Masking Technology},
booktitle={Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI},
year={2024},
pages={328-332},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012937600004508},
isbn={978-989-758-713-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI
TI - Virtual Try on Application and Parameter Analysis Based on Clothing Warping and Masking Technology
SN - 978-989-758-713-9
AU - Feng C.
PY - 2024
SP - 328
EP - 332
DO - 10.5220/0012937600004508
PB - SciTePress