Shen Fumin, Xu Yan, Liu Li, Yang Yang, Huang Zi, Shen Heng Tao
IEEE Trans Pattern Anal Mach Intell. 2018 Dec;40(12):3034-3044. doi: 10.1109/TPAMI.2018.2789887. Epub 2018 Jan 5.
Recent vision and learning studies show that learning compact hash codes can facilitate massive data processing with significantly reduced storage and computation. Particularly, learning deep hash functions has greatly improved the retrieval performance, typically under the semantic supervision. In contrast, current unsupervised deep hashing algorithms can hardly achieve satisfactory performance due to either the relaxed optimization or absence of similarity-sensitive objective. In this work, we propose a simple yet effective unsupervised hashing framework, named Similarity-Adaptive Deep Hashing (SADH), which alternatingly proceeds over three training modules: deep hash model training, similarity graph updating and binary code optimization. The key difference from the widely-used two-step hashing method is that the output representations of the learned deep model help update the similarity graph matrix, which is then used to improve the subsequent code optimization. In addition, for producing high-quality binary codes, we devise an effective discrete optimization algorithm which can directly handle the binary constraints with a general hashing loss. Extensive experiments validate the efficacy of SADH, which consistently outperforms the state-of-the-arts by large gaps.
近期的视觉与学习研究表明，学习紧凑哈希码能够显著减少存储和计算量，从而便于处理海量数据。特别地，学习深度哈希函数极大地提升了检索性能，通常是在语义监督下。相比之下，当前的无监督深度哈希算法由于优化松弛或缺乏相似性敏感目标，很难取得令人满意的性能。在这项工作中，我们提出了一个简单而有效的无监督哈希框架，名为相似性自适应深度哈希（SADH），它在三个训练模块上交替进行：深度哈希模型训练、相似性图更新和二进制代码优化。与广泛使用的两步哈希方法的关键区别在于，所学习的深度模型的输出表示有助于更新相似性图矩阵，然后该矩阵用于改进后续的代码优化。此外，为了生成高质量的二进制代码，我们设计了一种有效的离散优化算法，该算法可以直接处理具有一般哈希损失的二进制约束。大量实验验证了SADH的有效性，它始终大幅优于当前的先进方法。
