Learning Fine-grained Image Similarity with Deep Ranking

TL;DR

This paper introduces a deep ranking model with a multiscale network for fine-grained image similarity, outperforming traditional feature-based and classification models in capturing subtle differences.

Contribution

It presents a novel deep ranking approach with a multiscale network and an efficient triplet sampling algorithm for improved fine-grained image similarity learning.

Findings

Outperforms hand-crafted feature models in similarity tasks

Surpasses deep classification models in accuracy

Effective in capturing subtle image differences

Abstract

Learning fine-grained image similarity is a challenging task. It needs to capture between-class and within-class image differences. This paper proposes a deep ranking model that employs deep learning techniques to learn similarity metric directly from images.It has higher learning capability than models based on hand-crafted features. A novel multiscale network structure has been developed to describe the images effectively. An efficient triplet sampling algorithm is proposed to learn the model with distributed asynchronized stochastic gradient. Extensive experiments show that the proposed algorithm outperforms models based on hand-crafted visual features and deep classification models.