Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing
Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing
复制标题
通过基于概率的语义保留散列进行跨视图检索
DOI：
10.1109/tcyb.2016.2608906
复制
发表时间：
2017-12-01
影响因子：
11.8
通讯作者：
Wang, Jianmin
中科院分区：
文献类型：
--
作者：
Lin, Zijia;Ding, Guiguang;Wang, Jianmin
为了有效地从大规模多视图数据中检索最近邻，最近散列方法得到了广泛的研究，它可以大大提高查询速度。在本文中，我们提出了一个有效的基于概率的语义保持哈希（SePH）方法来解决跨视图检索的问题。考虑到视图之间的语义一致性，SePH为任何实例的所有观察视图生成一个统一的哈希码。对于训练，SePH首先将训练数据的给定语义亲和力转换为概率分布，并旨在通过最小化其Kullback-Leibler散度来近似Hamming空间中的另一个概率分布。具体地，后一种概率分布是从训练数据的待学习散列码之间的所有成对汉明距离导出的。然后，利用学习的散列码，任何类型的预测模型，如线性岭回归、逻辑回归或核逻辑回归，都可以被学习为每个视图中的散列函数，用于将对应的视图特定特征投影到散列码中。至于样本外扩展，给定任何未见过的实例，其观察到的视图中的学习散列函数可以预测视图特定的散列码。然后，通过导出或估计相对于预测的视图特定散列码的对应输出概率，进一步提出了一种新颖的概率方法来利用它们来确定统一散列码。为了评估所提出的SePH，我们在不同的基准数据集上进行了广泛的实验，实验结果表明，SePH是合理和有效的。
For efficiently retrieving nearest neighbors from large-scale multiview data, recently hashing methods are widely investigated, which can substantially improve query speeds. In this paper, we propose an effective probability-based semantics-preserving hashing (SePH) method to tackle the problem of cross-view retrieval. Considering the semantic consistency between views, SePH generates one unified hash code for all observed views of any instance. For training, SePH first transforms the given semantic affinities of training data into a probability distribution, and aims to approximate it with another one in Hamming space, via minimizing their Kullback–Leibler divergence. Specifically, the latter probability distribution is derived from all pair-wise Hamming distances between to-be-learnt hash codes of the training data. Then with learnt hash codes, any kind of predictive models like linear ridge regression, logistic regression, or kernel logistic regression, can be learnt as hash functions in each view for projecting the corresponding view-specific features into hash codes. As for out-of-sample extension, given any unseen instance, the learnt hash functions in its observed views can predict view-specific hash codes. Then by deriving or estimating the corresponding output probabilities with respect to the predicted view-specific hash codes, a novel probabilistic approach is further proposed to utilize them for determining a unified hash code. To evaluate the proposed SePH, we conduct extensive experiments on diverse benchmark datasets, and the experimental results demonstrate that SePH is reasonable and effective.