English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Cross-modal hashing aims to map heterogeneous multimedia data into a commonHamming space, which can realize fast and flexible retrieval across differentmodalities. Supervised cross-modal hashing methods have achieved considerableprogress by incorporating semantic side information. However, they mainly havetwo limitations: (1) Heavily rely on large-scale labeled cross-modal trainingdata which are labor intensive and hard to obtain. (2) Ignore the richinformation contained in the large amount of unlabeled data across differentmodalities, especially the margin examples that are easily to be incorrectlyretrieved, which can help to model the correlations. To address these problems,in this paper we propose a novel Semi-supervised Cross-Modal Hashing approachby Generative Adversarial Network (SCH-GAN). We aim to take advantage of GAN'sability for modeling data distributions to promote cross-modal hashing learningin an adversarial way. The main contributions can be summarized as follows: (1)We propose a novel generative adversarial network for cross-modal hashing. Inour proposed SCH-GAN, the generative model tries to select margin examples ofone modality from unlabeled data when giving a query of another modality. Whilethe discriminative model tries to distinguish the selected examples and truepositive examples of the query. These two models play a minimax game so thatthe generative model can promote the hashing performance of discriminativemodel. (2) We propose a reinforcement learning based algorithm to drive thetraining of proposed SCH-GAN. The generative model takes the correlation scorepredicted by discriminative model as a reward, and tries to select the examplesclose to the margin to promote discriminative model by maximizing the marginbetween positive and negative data. Experiments on 3 widely-used datasetsverify the effectiveness of our proposed approach.

SCH-GAN: Semi-supervised Cross-modal Hashing by Generative Adversarial Network