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

Generative adversarial network (GAN) has greatly improved the quality ofunsupervised image generation. Previous GAN-based methods often require a largeamount of high-quality training data while producing a small number (e.g.,tens) of classes. This work aims to scale up GANs to thousands of classesmeanwhile reducing the use of high-quality data in training. We propose animage generation method based on conditional transferring features, which cancapture pixel-level semantic changes when transforming low-quality images intohigh-quality ones. Moreover, self-supervision learning is integrated into ourGAN architecture to provide more label-free semantic supervisory informationobserved from the training data. As such, training our GAN architecturerequires much fewer high-quality images with a small number of additionallow-quality images. The experiments on CIFAR-10 and STL-10 show that evenremoving 30% high-quality images from the training set, our method can stilloutperform previous ones. The scalability on object classes has beenexperimentally validated: our method with 30% fewer high-quality images obtainsthe best quality in generating 1,000 ImageNet classes, as well as generatingall 3,755 classes of CASIA-HWDB1.0 Chinese handwriting characters.

Conditional Transferring Features: Scaling GANs to Thousands of Classes with 30% Less High-quality Data for Training