Hash Function Learning via Codewords
This work addresses hash learning for image retrieval, offering a flexible approach for supervised, unsupervised, and semi-supervised tasks, but it appears incremental as it builds on existing hash learning methods.
The paper tackles hash learning for content-based image retrieval by introducing a framework that uses inferred codewords in Hamming space to capture data similarity, achieving performance advantages as shown in comparative experiments.
In this paper we introduce a novel hash learning framework that has two main distinguishing features, when compared to past approaches. First, it utilizes codewords in the Hamming space as ancillary means to accomplish its hash learning task. These codewords, which are inferred from the data, attempt to capture similarity aspects of the data's hash codes. Secondly and more importantly, the same framework is capable of addressing supervised, unsupervised and, even, semi-supervised hash learning tasks in a natural manner. A series of comparative experiments focused on content-based image retrieval highlights its performance advantages.