Learning a Unified Embedding for Visual Search at Pinterest

TL;DR

This paper presents a multi-task deep metric learning system that creates a single unified image embedding for Pinterest's visual search, improving relevance, engagement, and operational efficiency across multiple applications.

Contribution

The work introduces a unified embedding approach that combines multiple visual search objectives into one deep neural network, outperforming specialized embeddings and reducing maintenance costs.

Findings

Unified embedding outperforms specialized embeddings in relevance and engagement.

Joint training on diverse data sources enhances embedding quality.

Binarized embeddings maintain high precision and recall for efficient retrieval.

Abstract

At Pinterest, we utilize image embeddings throughout our search and recommendation systems to help our users navigate through visual content by powering experiences like browsing of related content and searching for exact products for shopping. In this work we describe a multi-task deep metric learning system to learn a single unified image embedding which can be used to power our multiple visual search products. The solution we present not only allows us to train for multiple application objectives in a single deep neural network architecture, but takes advantage of correlated information in the combination of all training data from each application to generate a unified embedding that outperforms all specialized embeddings previously deployed for each product. We discuss the challenges of handling images from different domains such as camera photos, high quality web images, and clean product catalog images. We also detail how to jointly train for multiple product objectives and how to leverage both engagement data and human labeled data. In addition, our trained embeddings can also be binarized for efficient storage and retrieval without compromising precision and recall. Through comprehensive evaluations on offline metrics, user studies, and online A/B experiments, we demonstrate that our proposed unified embedding improves both relevance and engagement of our visual search products for both browsing and searching purposes when compared to existing specialized embeddings. Finally, the deployment of the unified embedding at Pinterest has drastically reduced the operation and engineering cost of maintaining multiple embeddings while improving quality.