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

Deep generative models have been successfully applied to many applications.However, existing works experience limitations when generating large images(the literature usually generates small images, e.g. 32 * 32 or 128 * 128). Inthis paper, we propose a novel scheme, called deep tensor adversarialgenerative nets (TGAN), that generates large high-quality images by exploringtensor structures. Essentially, the adversarial process of TGAN takes place ina tensor space. First, we impose tensor structures for concise imagerepresentation, which is superior in capturing the pixel proximity informationand the spatial patterns of elementary objects in images, over thevectorization preprocess in existing works. Secondly, we propose TGAN thatintegrates deep convolutional generative adversarial networks and tensorsuper-resolution in a cascading manner, to generate high-quality images fromrandom distributions. More specifically, we design a tensor super-resolutionprocess that consists of tensor dictionary learning and tensor coefficientslearning. Finally, on three datasets, the proposed TGAN generates images withmore realistic textures, compared with state-of-the-art adversarialautoencoders. The size of the generated images is increased by over 8.5 times,namely 374 * 374 in PASCAL2.

TGAN: Deep Tensor Generative Adversarial Nets for Large Image Generation