Fine-Grained Fashion Similarity Learning by Attribute-Specific Embedding Network

TL;DR

This paper introduces an Attribute-Specific Embedding Network (ASEN) that learns fine-grained fashion similarity by focusing on specific attributes, utilizing attention modules to improve the localization and pattern recognition for better similarity measurement.

Contribution

The paper proposes a novel end-to-end network with attention modules to learn multiple attribute-specific embeddings for fine-grained fashion similarity, advancing beyond existing methods.

Findings

ASEN outperforms baseline models on four fashion datasets.

Attention modules improve attribute localization and pattern capturing.

Enhanced fine-grained similarity measurement benefits fashion reranking applications.

Abstract

This paper strives to learn fine-grained fashion similarity. In this similarity paradigm, one should pay more attention to the similarity in terms of a specific design/attribute among fashion items, which has potential values in many fashion related applications such as fashion copyright protection. To this end, we propose an Attribute-Specific Embedding Network (ASEN) to jointly learn multiple attribute-specific embeddings in an end-to-end manner, thus measure the fine-grained similarity in the corresponding space. With two attention modules, i.e., Attribute-aware Spatial Attention and Attribute-aware Channel Attention, ASEN is able to locate the related regions and capture the essential patterns under the guidance of the specified attribute, thus make the learned attribute-specific embeddings better reflect the fine-grained similarity. Extensive experiments on four fashion-related datasets show the effectiveness of ASEN for fine-grained fashion similarity learning and its potential for fashion reranking.