Search papers, labs, and topics across Lattice.
This paper introduces SuperFashion, a novel framework for Attribute-Specific Fashion Retrieval (ASFR) that utilizes superpixel tokens within a Transformer architecture to enhance fine-grained image retrieval. By employing an attribute-guided attention mechanism and superpixel segmentation, SuperFashion effectively captures subtle pixel-level microstructures while minimizing background noise. Experimental results on multiple datasets show significant improvements in mean average precision (MAP), with gains of up to 9.35% over state-of-the-art methods, highlighting its effectiveness in attribute localization and discrimination.
SuperFashion achieves up to 9.35% improvement in fashion image retrieval by leveraging superpixel tokens for enhanced attribute localization.
Attribute-Specific Fashion Retrieval (ASFR) aims to improve fine-grained image retrieval by focusing on specific attributes. However, existing patch-based attention and Transformer methods often misalign with irregular attribute regions and are prone to background noise, limiting their ability to capture subtle, pixel-level microstructures. To tackle these challenges, we propose SuperFashion, the first ASFR framework that adopts superpixel tokens within a Transformer architecture. SuperFashion initially employs an attribute-guided attention mechanism to extract attribute-related features, which in turn guide the cropping of semantically meaningful image regions. Superpixel segmentation is then leveraged on these regions to generate compact, semantically coherent superpixel tokens. By incorporating modality-specific embeddings for both attribute and superpixel tokens, the superpixel token-based Transformer facilitates adaptive interaction and fusion, thereby enhancing attribute localization and discrimination. Extensive experiments on FashionAI, DARN, and DeepFashion demonstrate relative overall MAP improvements of 1.84%, 9.27%, and 9.35% over prior SOTA. SuperFashion offers a new solution for web-based image retrieval.
