Leveraging Weakly Annotated Data for Fashion Image Retrieval and Label Prediction

Abstract

In this paper, we present a method to learn a visual representation adapted for e-commerce products. Based on weakly supervised learning, our model learns from noisy datasets crawled on e-commerce website catalogs and does not require any manual labeling. We show that our representation can be used for downward classification tasks over clothing categories with different levels of granularity. We also demonstrate that the learnt representation is suitable for image retrieval. We achieve nearly state-of-art results on the DeepFashion In-Shop Clothes Retrieval and Categories Attributes Prediction tasks, without using the provided training set.

References in corpus (4)

• NLTK: The Natural Language Toolkit
• Deep Learning based Large Scale Visual Recommendation and Search for E-Commerce
• Hard-Aware Deeply Cascaded Embedding
• Fully-adaptive Feature Sharing in Multi-Task Networks with Applications in Person Attribute Classification

Cited by in corpus (14)

• Fashion Meets Computer Vision: A Survey
• FINE Samples for Learning with Noisy Labels
• Automated System for Ship Detection from Medium Resolution Satellite Optical Imagery
• Fashion Retrieval via Graph Reasoning Networks on a Similarity Pyramid
• A Unified Model with Structured Output for Fashion Images Classification
• Limb-Aware Virtual Try-On Network with Progressive Clothing Warping
• Semi-supervised Feature-Level Attribute Manipulation for Fashion Image Retrieval
• Progressive Limb-Aware Virtual Try-On
• Fashion Image Retrieval with Capsule Networks
• Two-Stream Multi-Task Network for Fashion Recognition
• Product1M: Towards Weakly Supervised Instance-Level Product Retrieval via Cross-modal Pretraining
• Semi-supervised Learning with a Teacher-student Network for Generalized Attribute Prediction
• Hard-Aware Fashion Attribute Classification
• FashionSearchNet-v2: Learning Attribute Representations with Localization for Image Retrieval with Attribute Manipulation