Studio2Shop: from studio photo shoots to fashion articles

TL;DR

This paper introduces Studio2Shop, a deep learning model designed to match full-body or half-body studio images of clothing with corresponding fashion articles, advancing visual search capabilities in fashion retrieval.

Contribution

The paper presents a novel deep convolutional network that leverages domain-specific numerical representations for accurate studio-to-shop fashion item matching.

Findings

High accuracy in top-$k$ retrieval results

Effective matching of studio images to shop articles

Potential for realistic visual search engine development

Abstract

Fashion is an increasingly important topic in computer vision, in particular the so-called street-to-shop task of matching street images with shop images containing similar fashion items. Solving this problem promises new means of making fashion searchable and helping shoppers find the articles they are looking for. This paper focuses on finding pieces of clothing worn by a person in full-body or half-body images with neutral backgrounds. Such images are ubiquitous on the web and in fashion blogs, and are typically studio photos, we refer to this setting as studio-to-shop. Recent advances in computational fashion include the development of domain-specific numerical representations. Our model Studio2Shop builds on top of such representations and uses a deep convolutional network trained to match a query image to the numerical feature vectors of all the articles annotated in this image. Top-$k$ retrieval evaluation on test query images shows that the correct items are most often found within a range that is sufficiently small for building realistic visual search engines for the studio-to-shop setting.