FashionFAE: Fine-grained Attributes Enhanced Fashion Vision-Language Pre-training

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Jiale, Gao, Dehong, Zhang, Jinxia, Zhan, Zechao, Hu, Yang, Wang, Xin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929671085490176
author Huang, Jiale
Gao, Dehong
Zhang, Jinxia
Zhan, Zechao
Hu, Yang
Wang, Xin
author_facet Huang, Jiale
Gao, Dehong
Zhang, Jinxia
Zhan, Zechao
Hu, Yang
Wang, Xin
contents Large-scale Vision-Language Pre-training (VLP) has demonstrated remarkable success in the general domain. However, in the fashion domain, items are distinguished by fine-grained attributes like texture and material, which are crucial for tasks such as retrieval. Existing models often fail to leverage these fine-grained attributes from both text and image modalities. To address the above issues, we propose a novel approach for the fashion domain, Fine-grained Attributes Enhanced VLP (FashionFAE), which focuses on the detailed characteristics of fashion data. An attribute-emphasized text prediction task is proposed to predict fine-grained attributes of the items. This forces the model to focus on the salient attributes from the text modality. Additionally, a novel attribute-promoted image reconstruction task is proposed, which further enhances the fine-grained ability of the model by leveraging the representative attributes from the image modality. Extensive experiments show that FashionFAE significantly outperforms State-Of-The-Art (SOTA) methods, achieving 2.9% and 5.2% improvements in retrieval on sub-test and full test sets, respectively, and a 1.6% average improvement in recognition tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2412_19997
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FashionFAE: Fine-grained Attributes Enhanced Fashion Vision-Language Pre-training
Huang, Jiale
Gao, Dehong
Zhang, Jinxia
Zhan, Zechao
Hu, Yang
Wang, Xin
Computer Vision and Pattern Recognition
Large-scale Vision-Language Pre-training (VLP) has demonstrated remarkable success in the general domain. However, in the fashion domain, items are distinguished by fine-grained attributes like texture and material, which are crucial for tasks such as retrieval. Existing models often fail to leverage these fine-grained attributes from both text and image modalities. To address the above issues, we propose a novel approach for the fashion domain, Fine-grained Attributes Enhanced VLP (FashionFAE), which focuses on the detailed characteristics of fashion data. An attribute-emphasized text prediction task is proposed to predict fine-grained attributes of the items. This forces the model to focus on the salient attributes from the text modality. Additionally, a novel attribute-promoted image reconstruction task is proposed, which further enhances the fine-grained ability of the model by leveraging the representative attributes from the image modality. Extensive experiments show that FashionFAE significantly outperforms State-Of-The-Art (SOTA) methods, achieving 2.9% and 5.2% improvements in retrieval on sub-test and full test sets, respectively, and a 1.6% average improvement in recognition tasks.
title FashionFAE: Fine-grained Attributes Enhanced Fashion Vision-Language Pre-training
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.19997