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

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. Code and data are available at https://github.com/Maryeon/asen.

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

“…A weakly-supervised localization method and a two-stage training strategy are also devised to fulfil the two-branch model. Experimentally, compared to the ASEN with only one branch proposed in [23], the new model with two branches consistently leads to better fine-grained fashion retrieval performance on three datasets. Besides, more extensive experiments are included: exploring the viability of each component, the influence of the hyperparameters in ASEN.…”

“…• ASEN g [23]: It is the conference version of our ASEN, where only the global branch is employed to learn the attribute-specific embeddings.…”

“…• Extensive experiments on FashionAI [22], DARN [6] and DeepFashion [1] datasets demonstrate the effectiveness of ASEN for fine-grained fashion similarity prediction and its potential for fashion reranking. A preliminary version of this work was published at AAAI 2020 [23]. The journal extension mainly improves over the conference paper in following aspects.…”

Hierarchical Similarity Learning for Language-Based Product Image Retrieval

“…Cross-modal retrieval is a classical task at the intersection between computer vision and natural language processing, and has been widely explored [1,2,3,4]. Recently, with the increasing popularity of e-commerce platforms [5,6,7,8], language-based product image retrieval attracts increasing attention [9,10,11]. As exemplified in Fig.…”

Multi-Order Adversarial Representation Learning for Composed Query Image Retrieval

“…Image retrieval is an important research topic in the field of computer vision and multimedia, and has a wide range of applications [1,2], such as fashion retrieval [3,4] and person re-identification [5,6]. The retrieval system usually only uses a picture or a paragraph of text as input.…”