Column Sampling Based Discrete Supervised Hashing
Column Sampling Based Discrete Supervised Hashing
复制标题
DOI：
10.1609/aaai.v30i1.10176
复制
发表时间：
2016-02
期刊：
影响因子：
--
通讯作者：
Wang-Cheng Kang;Wu-Jun Li;Zhi-Hua Zhou
中科院分区：
文献类型：
--
作者：
Wang-Cheng Kang;Wu-Jun Li;Zhi-Hua Zhou
通过利用语义（标签）信息，监督哈希在许多真实的应用中表现出比无监督哈希更好的准确性。由于散列码学习问题本质上是一个难以求解的离散优化问题，大多数现有的监督散列方法试图通过丢弃离散约束来解决松弛的连续优化问题。然而，这些方法通常由于松弛引起的误差而遭受较差的性能。其他一些方法试图直接解决离散优化问题。然而，它们通常是耗时且不可扩展的。在本文中，我们提出了一种新的方法，称为基于列采样离散监督哈希（COSDISH），直接学习离散哈希代码的语义信息。COSDISH是一种迭代方法，在每次迭代中，从语义相似度矩阵中采样几列，然后将散列代码分解为两部分，这两部分可以以离散方式交替优化。理论分析表明，COSDISH的学习（优化）算法在交替优化过程的每一步都有一个恒定的近似界。在具有语义标签的数据集上的实验结果表明，COSDISH在图像检索等真实的应用中的性能优于最先进的方法。
By leveraging semantic (label) information, supervised hashing has demonstrated better accuracy than unsupervised hashing in many real applications. Because the hashing-code learning problem is essentially a discrete optimization problem which is hard to solve, most existing supervised hashing methods try to solve a relaxed continuous optimization problem by dropping the discrete constraints. However, these methods typically suffer from poor performance due to the errors caused by the relaxation. Some other methods try to directly solve the discrete optimization problem. However, they are typically time-consuming and unscalable. In this paper, we propose a novel method, called column sampling based discrete supervised hashing (COSDISH), to directly learn the discrete hashing code from semantic information. COSDISH is an iterative method, in each iteration of which several columns are sampled from the semantic similarity matrix and then the hashing code is decomposed into two parts which can be alternately optimized in a discrete way. Theoretical analysis shows that the learning (optimization) algorithm of COSDISH has a constant-approximation bound in each step of the alternating optimization procedure. Empirical results on datasets with semantic labels illustrate that COSDISH can outperform the state-of-the-art methods in real applications like image retrieval.