HHF: Hashing-guided Hinge Function for Deep Hashing Retrieval

TL;DR

This paper introduces HHF, a novel hashing-guided hinge function that balances metric learning and quantization in deep hashing, significantly improving image retrieval accuracy across multiple datasets.

Contribution

The paper proposes a new HHF method that explicitly balances metric and quantization terms, overcoming conflicts in deep hashing and enhancing retrieval performance.

Findings

HHF outperforms existing deep hashing methods on CIFAR-10, CIFAR-100, ImageNet, and MS-COCO.

The method is robust and adaptable to other deep hashing techniques.

Extensive experiments validate the effectiveness of HHF in large-scale image retrieval.

Abstract

Deep hashing has shown promising performance in large-scale image retrieval. However, latent codes extracted by Deep Neural Networks (DNNs) will inevitably lose semantic information during the binarization process, which damages the retrieval accuracy and makes it challenging. Although many existing approaches perform regularization to alleviate quantization errors, we figure out an incompatible conflict between metric learning and quantization learning. The metric loss penalizes the inter-class distances to push different classes unconstrained far away. Worse still, it tends to map the latent code deviate from ideal binarization point and generate severe ambiguity in the binarization process. Based on the minimum distance of the binary linear code, we creatively propose Hashing-guided Hinge Function (HHF) to avoid such conflict. In detail, the carefully-designed inflection point, which relies on the hash bit length and category numbers, is explicitly adopted to balance the metric term and quantization term. Such a modification prevents the network from falling into local metric optimal minima in deep hashing. Extensive experiments in CIFAR-10, CIFAR-100, ImageNet, and MS-COCO show that HHF consistently outperforms existing techniques, and is robust and flexible to transplant into other methods. Code is available at https://github.com/JerryXu0129/HHF.