Fashion-Gen: The Generative Fashion Dataset and Challenge

Abstract

We introduce a new dataset of 293,008 high definition (1360 x 1360 pixels) fashion images paired with item descriptions provided by professional stylists. Each item is photographed from a variety of angles. We provide baseline results on 1) high-resolution image generation, and 2) image generation conditioned on the given text descriptions. We invite the community to improve upon these baselines. In this paper, we also outline the details of a challenge that we are launching based upon this dataset.

References in corpus (14)

• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
• Generative Adversarial Text to Image Synthesis
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Improved Techniques for Training GANs
• Unsupervised Cross-Domain Image Generation
• Stochastic Video Generation with a Learned Prior
• A Note on the Inception Score
• Unsupervised Learning of Disentangled Representations from Video
• Amortised MAP Inference for Image Super-resolution
• Photographic Text-to-Image Synthesis with a Hierarchically-nested Adversarial Network
• Be Your Own Prada: Fashion Synthesis with Structural Coherence
• Learning Visual Clothing Style with Heterogeneous Dyadic Co-occurrences
• VITON: An Image-based Virtual Try-on Network
• Hierarchical Adversarially Learned Inference

Cited by in corpus (17)

• VLP: A Survey on Vision-Language Pre-training
• Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback
• CurlingNet: Compositional Learning between Images and Text for Fashion IQ Data
• Compatible and Diverse Fashion Image Inpainting
• Kaleido-BERT: Vision-Language Pre-training on Fashion Domain
• Context-Aware Visual Compatibility Prediction
• Fashion-IQ 2020 Challenge 2nd Place Team's Solution
• Approximating Human Judgment of Generated Image Quality
• Exploiting Latent Codes: Interactive Fashion Product Generation, Similar Image Retrieval, and Cross-Category Recommendation using Variational Autoencoders
• Self-supervised Visual Attribute Learning for Fashion Compatibility
• Towards Co-Creative Generative Adversarial Networks for Fashion Designers
• TailorGAN: Making User-Defined Fashion Designs
• Multi-Modality Transformer for E-Commerce: Inferring User Purchase Intention to Bridge the Query-Product Gap
• Fashion is Taking Shape: Understanding Clothing Preference Based on Body Shape From Online Sources
• PatentNet: A Large-Scale Incomplete Multiview, Multimodal, Multilabel Industrial Goods Image Database
• Fashion-AttGAN: Attribute-Aware Fashion Editing with Multi-Objective GAN
• Conversational Fashion Image Retrieval via Multiturn Natural Language Feedback