Regularizing Deep Hashing Networks Using GAN Generated Fake Images

TL;DR

This paper introduces a novel regularization technique for deep hashing networks using GAN-generated fake images, improving image retrieval performance by reducing overfitting.

Contribution

The paper proposes a two-stage training pipeline that uses GAN-generated fake images and a maximum-entropy loss to regularize deep hashing models, a novel approach in deep hashing.

Findings

Achieves superior retrieval accuracy on benchmark datasets.

Demonstrates effective regularization reduces overfitting.

Outperforms state-of-the-art hashing methods.

Abstract

Recently, deep-networks-based hashing (deep hashing) has become a leading approach for large-scale image retrieval. It aims to learn a compact bitwise representation for images via deep networks, so that similar images are mapped to nearby hash codes. Since a deep network model usually has a large number of parameters, it may probably be too complicated for the training data we have, leading to model over-fitting. To address this issue, in this paper, we propose a simple two-stage pipeline to learn deep hashing models, by regularizing the deep hashing networks using fake images. The first stage is to generate fake images from the original training set without extra data, via a generative adversarial network (GAN). In the second stage, we propose a deep architec- ture to learn hash functions, in which we use a maximum-entropy based loss to incorporate the newly created fake images by the GAN. We show that this loss acts as a strong regularizer of the deep architecture, by penalizing low-entropy output hash codes. This loss can also be interpreted as a model ensemble by simultaneously training many network models with massive weight sharing but over different training sets. Empirical evaluation results on several benchmark datasets show that the proposed method has superior performance gains over state-of-the-art hashing methods.