Hierarchical Average Precision Training for Pertinent Image Retrieval

TL;DR

This paper introduces HAPPIER, a hierarchical Average Precision training method for image retrieval that incorporates a concept hierarchy to improve ranking evaluation and model training, outperforming existing methods.

Contribution

The paper proposes a novel hierarchical AP metric and a training method that leverages this metric to enhance pertinent image retrieval performance.

Findings

HAPPIER outperforms state-of-the-art hierarchical retrieval methods.

HAPPIER matches the best results on fine-grained ranking tasks.

It improves embedding space organization and reduces severe failure cases.

Abstract

Image Retrieval is commonly evaluated with Average Precision (AP) or Recall@k. Yet, those metrics, are limited to binary labels and do not take into account errors' severity. This paper introduces a new hierarchical AP training method for pertinent image retrieval (HAP-PIER). HAPPIER is based on a new H-AP metric, which leverages a concept hierarchy to refine AP by integrating errors' importance and better evaluate rankings. To train deep models with H-AP, we carefully study the problem's structure and design a smooth lower bound surrogate combined with a clustering loss that ensures consistent ordering. Extensive experiments on 6 datasets show that HAPPIER significantly outperforms state-of-the-art methods for hierarchical retrieval, while being on par with the latest approaches when evaluating fine-grained ranking performances. Finally, we show that HAPPIER leads to better organization of the embedding space, and prevents most severe failure cases of non-hierarchical methods. Our code is publicly available at: https://github.com/elias-ramzi/HAPPIER.