Supervised Metric Learning to Rank for Retrieval via Contextual Similarity Optimization

TL;DR

This paper introduces a new metric learning method called contextual loss that improves image retrieval by enhancing robustness to label noise and reducing overfitting, achieving state-of-the-art results.

Contribution

It proposes a novel contextual loss function that optimizes semantic consistency and improves retrieval performance over existing methods.

Findings

More robust to label noise

Less prone to overfitting

Achieves state-of-the-art results on four benchmarks

Abstract

There is extensive interest in metric learning methods for image retrieval. Many metric learning loss functions focus on learning a correct ranking of training samples, but strongly overfit semantically inconsistent labels and require a large amount of data. To address these shortcomings, we propose a new metric learning method, called contextual loss, which optimizes contextual similarity in addition to cosine similarity. Our contextual loss implicitly enforces semantic consistency among neighbors while converging to the correct ranking. We empirically show that the proposed loss is more robust to label noise, and is less prone to overfitting even when a large portion of train data is withheld. Extensive experiments demonstrate that our method achieves a new state-of-the-art across four image retrieval benchmarks and multiple different evaluation settings. Code is available at: https://github.com/Chris210634/metric-learning-using-contextual-similarity