M

Marqo Fashionsiglip ST

Developed by pySilver
Marqo-FashionSigLIP is a multimodal embedding model optimized for fashion product search, achieving a 57% improvement in MRR and recall rate compared to FashionCLIP.
Downloads 3,586
Release Time : 3/3/2025

Model Overview

This model is fine-tuned based on ViT-B-16-SigLIP (webli) and trained using Generalized Contrastive Learning (GCL). It can provide highly relevant search results for fashion products through text descriptions, categories, styles, colors, and more.

Model Features

Multimodal Retrieval
Supports multimodal retrieval of fashion products through text and images.
Generalized Contrastive Learning
Utilizes GCL for training with multiple features such as categories, styles, and colors.
High Performance
Outperforms similar models on multiple fashion datasets with significant improvements in recall rate and MRR.

Model Capabilities

Zero-shot image classification
Multimodal retrieval
Fashion product search
Text-to-image retrieval
Image-to-text retrieval

Use Cases

E-commerce
Fashion Product Search
Search for related fashion products via text descriptions or images.
57% improvement in recall rate and MRR compared to FashionCLIP.
Fashion Recommendation
Personalized Recommendation
Recommend related fashion products based on user input or browsing history.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase