Attribute-Guided Multi-Level Attention Network for Fine-Grained Fashion Retrieval

TL;DR

This paper introduces AG-MAN, a novel attribute-guided multi-level attention network that enhances feature extraction for fine-grained fashion retrieval, addressing the feature gap problem and outperforming existing methods on multiple datasets.

Contribution

The paper proposes a new attribute-guided multi-level attention network with enhanced feature extraction and a classification scheme to improve fine-grained fashion retrieval accuracy.

Findings

Outperforms existing attention-based methods on FashionAI, DeepFashion, and Zappos50k datasets.

Achieves up to 2.12% improvement in MAP on FashionAI dataset.

Effectively alleviates the feature gap problem in fine-grained fashion retrieval.

Abstract

Fine-grained fashion retrieval searches for items that share a similar attribute with the query image. Most existing methods use a pre-trained feature extractor (e.g., ResNet 50) to capture image representations. However, a pre-trained feature backbone is typically trained for image classification and object detection, which are fundamentally different tasks from fine-grained fashion retrieval. Therefore, existing methods suffer from a feature gap problem when directly using the pre-trained backbone for fine-tuning. To solve this problem, we introduce an attribute-guided multi-level attention network (AG-MAN). Specifically, we first enhance the pre-trained feature extractor to capture multi-level image embedding, thereby enriching the low-level features within these representations. Then, we propose a classification scheme where images with the same attribute, albeit with different values, are categorized into the same class. This can further alleviate the feature gap problem by perturbing object-centric feature learning. Moreover, we propose an improved attribute-guided attention module for extracting more accurate attribute-specific representations. Our model consistently outperforms existing attention based methods when assessed on the FashionAI (62.8788% in MAP), DeepFashion (8.9804% in MAP), and Zappos50k datasets (93.32% in Prediction accuracy). Especially, ours improves the most typical ASENet_V2 model by 2.12%, 0.31%, and 0.78% points in FashionAI, DeepFashion, and Zappos50k datasets, respectively. The source code is available in https://github.com/Dr-LingXiao/AG-MAN.