[论文解读] Deep Hashing Network for Unsupervised Domain Adaptation

Hemanth Venkateswara, José Eusébio+2 more|arXiv (Cornell University)|Jun 22, 2017

Advanced Image and Video Retrieval Techniques被引用 6

一句话总结

本文提出领域自适应哈希（DAH），一种深度学习框架，通过联合学习紧凑的二进制哈希码并利用有标签的源域数据与无标签的目标域数据对齐领域分布，实现了哈希领域自适应的最先进性能。在新提出的Office-Home数据集上，DAH在64位哈希码长度下实现了0.480的平均mAP，优于现有方法。

ABSTRACT

In recent years, deep neural networks have emerged as a dominant machine learning tool for a wide variety of application domains. However, training a deep neural network requires a large amount of labeled data, which is an expensive process in terms of time, labor and human expertise. Domain adaptation or transfer learning algorithms address this challenge by leveraging labeled data in a different, but related source domain, to develop a model for the target domain. Further, the explosive growth of digital data has posed a fundamental challenge concerning its storage and retrieval. Due to its storage and retrieval efficiency, recent years have witnessed a wide application of hashing in a variety of computer vision applications. In this paper, we first introduce a new dataset, Office-Home, to evaluate domain adaptation algorithms. The dataset contains images of a variety of everyday objects from multiple domains. We then propose a novel deep learning framework that can exploit labeled source data and unlabeled target data to learn informative hash codes, to accurately classify unseen target data. To the best of our knowledge, this is the first research effort to exploit the feature learning capabilities of deep neural networks to learn representative hash codes to address the domain adaptation problem. Our extensive empirical studies on multiple transfer tasks corroborate the usefulness of the framework in learning efficient hash codes which outperform existing competitive baselines for unsupervised domain adaptation.

研究动机与目标

• 通过利用相关源域的有标签数据，解决目标域中标签数据有限的挑战。
• 开发一个统一的深度学习框架，同时学习信息丰富的哈希码并减少领域偏移。
• 通过使用紧凑的二进制表示，提升无监督领域自适应下图像分类的效率与准确性。
• 引入一个新的基准数据集Office-Home，用于评估物体识别中的领域自适应算法。

提出的方法

• 提出一种深度神经网络，输出二进制哈希码而非概率向量，实现高效存储与检索。
• 采用监督哈希损失，确保同类别源域样本具有相似的哈希码。
• 应用无监督熵损失，促使每个无标签目标样本与单一源类别对齐。
• 引入多核最大均值差异（MK-MMD）损失，以最小化网络各层之间的领域分布差异。
• 使用结合所有三个组件的复合损失函数，端到端训练网络。
• 使用包含65个类别和4个领域（艺术、剪贴画、产品、真实世界）共15,500张图像的Office-Home数据集进行评估。

Figure 1: The Domain Adaptive Hash (DAH) network that outputs hash codes for the source and the target. The network is trained with a batch of source and target data. The convolution layers conv 1 - conv 5 and the fully connected layers fc 6 and fc 7 are fine tuned from the VGG-F network. The MK-MMD

实验结果

研究问题

• RQ1 深度神经网络能否有效适应以学习用于无监督领域自适应的紧凑哈希码？
• RQ2 与分别训练相比，联合优化哈希码学习与领域对齐如何提升性能？
• RQ3 源域中的有标签数据在多大程度上能提升无标签目标域中哈希码的质量？
• RQ4 与现有无监督和有监督哈希方法相比，所提出的DAH框架在准确率和效率方面表现如何？

主要发现

• 所提出的DAH框架在Office-Home数据集上，64位哈希码长度下实现了0.480的平均平均精度（mAP），显著优于所有无监督基线方法。
• 在Office数据集上，DAH在Webcam上的mAP达到0.717，在Amazon上的mAP达到0.582，分别优于最佳无监督方法（BDNN）6.5%和12.6%。
• DAH始终优于NoDA（无领域自适应）方法，后者受领域偏移影响，证明了在哈希中进行领域对齐的必要性。
• 无监督熵损失能有效引导目标样本与源类别对齐，即使无标签，也能提升零样本设置下的鲁棒性。
• DAH性能接近使用目标标签的有监督哈希（SuH），表明其在弱监督场景下的有效性。
• 该框架在多种领域偏移场景下泛化良好，包括从艺术到剪贴画、产品到真实世界，以及Amazon到Webcam，展现出强大的可迁移性。

Figure 2: Sample images from the Office-Home dataset. The dataset consists of images of everyday objects organized into 4 domains; Art : paintings, sketches and/or artistic depictions, Clipart : clipart images, Product : images without background and Real-World : regular images captured with a camer