Data Augmentation using Random Image Cropping for High-resolution Virtual Try-On (VITON-CROP)

TL;DR

VITON-CROP introduces a novel data augmentation technique using random image cropping to improve the realism and robustness of high-resolution virtual try-on images, outperforming existing models.

Contribution

The paper presents VITON-CROP, a new method that enhances virtual try-on results through random crop augmentation, addressing limitations of previous approaches.

Findings

VITON-CROP produces more realistic virtual try-on images.

It outperforms VITON-HD both qualitatively and quantitatively.

Random cropping improves model robustness.

Abstract

Image-based virtual try-on provides the capacity to transfer a clothing item onto a photo of a given person, which is usually accomplished by warping the item to a given human pose and adjusting the warped item to the person. However, the results of real-world synthetic images (e.g., selfies) from the previous method is not realistic because of the limitations which result in the neck being misrepresented and significant changes to the style of the garment. To address these challenges, we propose a novel method to solve this unique issue, called VITON-CROP. VITON-CROP synthesizes images more robustly when integrated with random crop augmentation compared to the existing state-of-the-art virtual try-on models. In the experiments, we demonstrate that VITON-CROP is superior to VITON-HD both qualitatively and quantitatively.