Toward Accurate and Realistic Outfits Visualization with Attention to Details

TL;DR

This paper introduces OVNet, a novel virtual try-on system that captures detailed visual features like buttons and textures, producing realistic multi-garment images with an interactive interface for e-commerce.

Contribution

We propose a new outfit visualization network with a cascade warping approach and outfit-model matching, significantly improving detail accuracy and realism over prior methods.

Findings

Generated images show higher realism and detail quality.

Quantitative analysis confirms superior performance.

Deployed interface received positive user feedback.

Abstract

Virtual try-on methods aim to generate images of fashion models wearing arbitrary combinations of garments. This is a challenging task because the generated image must appear realistic and accurately display the interaction between garments. Prior works produce images that are filled with artifacts and fail to capture important visual details necessary for commercial applications. We propose Outfit Visualization Net (OVNet) to capture these important details (e.g. buttons, shading, textures, realistic hemlines, and interactions between garments) and produce high quality multiple-garment virtual try-on images. OVNet consists of 1) a semantic layout generator and 2) an image generation pipeline using multiple coordinated warps. We train the warper to output multiple warps using a cascade loss, which refines each successive warp to focus on poorly generated regions of a previous warp and yields consistent improvements in detail. In addition, we introduce a method for matching outfits with the most suitable model and produce significant improvements for both our and other previous try-on methods. Through quantitative and qualitative analysis, we demonstrate our method generates substantially higher-quality studio images compared to prior works for multi-garment outfits. An interactive interface powered by this method has been deployed on fashion e-commerce websites and received overwhelmingly positive feedback.