TGAN: Deep Tensor Generative Adversarial Nets for Large Image Generation
This work addresses a limitation in generative models for producing large images, which is important for applications requiring high-resolution visual content, but it appears incremental as it builds on existing GAN and tensor methods.
The paper tackles the problem of generating large high-quality images, which existing methods struggle with, by proposing TGAN, a deep tensor generative adversarial network that integrates tensor structures and super-resolution, resulting in images with more realistic textures and an 8.5 times increase in size to 374x374 on the PASCAL2 dataset.
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). In this paper, we propose a novel scheme, called deep tensor adversarial generative nets (TGAN), that generates large high-quality images by exploring tensor structures. Essentially, the adversarial process of TGAN takes place in a tensor space. First, we impose tensor structures for concise image representation, which is superior in capturing the pixel proximity information and the spatial patterns of elementary objects in images, over the vectorization preprocess in existing works. Secondly, we propose TGAN that integrates deep convolutional generative adversarial networks and tensor super-resolution in a cascading manner, to generate high-quality images from random distributions. More specifically, we design a tensor super-resolution process that consists of tensor dictionary learning and tensor coefficients learning. Finally, on three datasets, the proposed TGAN generates images with more realistic textures, compared with state-of-the-art adversarial autoencoders. The size of the generated images is increased by over 8.5 times, namely 374 * 374 in PASCAL2.