Dress Code: High-Resolution Multi-Category Virtual Try-On

TL;DR

This paper introduces Dress Code, a large, high-resolution dataset for multi-category virtual try-on, and proposes a semantic-aware discriminator to improve visual quality in try-on images.

Contribution

The paper presents a new high-resolution, multi-category dataset and a novel semantic-aware discriminator for enhanced virtual try-on performance.

Findings

Outperforms baselines in visual quality

Achieves state-of-the-art quantitative results

Enables detailed full-body virtual try-on

Abstract

Image-based virtual try-on strives to transfer the appearance of a clothing item onto the image of a target person. Prior work focuses mainly on upper-body clothes (e.g. t-shirts, shirts, and tops) and neglects full-body or lower-body items. This shortcoming arises from a main factor: current publicly available datasets for image-based virtual try-on do not account for this variety, thus limiting progress in the field. To address this deficiency, we introduce Dress Code, which contains images of multi-category clothes. Dress Code is more than 3x larger than publicly available datasets for image-based virtual try-on and features high-resolution paired images (1024x768) with front-view, full-body reference models. To generate HD try-on images with high visual quality and rich in details, we propose to learn fine-grained discriminating features. Specifically, we leverage a semantic-aware discriminator that makes predictions at pixel-level instead of image- or patch-level. Extensive experimental evaluation demonstrates that the proposed approach surpasses the baselines and state-of-the-art competitors in terms of visual quality and quantitative results. The Dress Code dataset is publicly available at https://github.com/aimagelab/dress-code.