Deep Attention-Guided Hashing
| Autor: | Jun Long, Osolo Ian Raymond, Zhan Yang, Wuqing Sun |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
FOS: Computer and information sciences
Theoretical computer science General Computer Science Multimedia search Computer science Hash function General Engineering 02 engineering and technology Function (mathematics) 010501 environmental sciences 01 natural sciences Computer Science - Information Retrieval Image (mathematics) attention-guided strategy Data_FILES 0202 electrical engineering electronic engineering information engineering Benchmark (computing) 020201 artificial intelligence & image processing General Materials Science Supervised learning-based hashing lcsh:Electrical engineering. Electronics. Nuclear engineering lcsh:TK1-9971 Information Retrieval (cs.IR) 0105 earth and related environmental sciences |
| Zdroj: | IEEE Access, Vol 7, Pp 11209-11221 (2019) |
| ISSN: | 2169-3536 |
| DOI: | 10.1109/access.2019.2891894 |
| Popis: | With the rapid growth of multimedia data (e.g., image, audio and video etc.) on the web, learning-based hashing techniques such as Deep Supervised Hashing (DSH) have proven to be very efficient for large-scale multimedia search. The recent successes seen in Learning-based hashing methods are largely due to the success of deep learning-based hashing methods. However, there are some limitations to previous learning-based hashing methods (e.g., the learned hash codes containing repetitive and highly correlated information). In this paper, we propose a novel learning-based hashing method, named Deep Attention-guided Hashing (DAgH). DAgH is implemented using two stream frameworks. The core idea is to use guided hash codes which are generated by the hashing network of the first stream framework (called first hashing network) to guide the training of the hashing network of the second stream framework (called second hashing network). Specifically, in the first network, it leverages an attention network and hashing network to generate the attention-guided hash codes from the original images. The loss function we propose contains two components: the semantic loss and the attention loss. The attention loss is used to punish the attention network to obtain the salient region from pairs of images; in the second network, these attention-guided hash codes are used to guide the training of the second hashing network (i.e., these codes are treated as supervised labels to train the second network). By doing this, DAgH can make full use of the most critical information contained in images to guide the second hashing network in order to learn efficient hash codes in a true end-to-end fashion. Results from our experiments demonstrate that DAgH can generate high quality hash codes and it outperforms current state-of-the-art methods on three benchmark datasets, CIFAR-10, NUS-WIDE, and ImageNet. Comment: Accepted to IEEE ACCESS |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|