Pseudo-Supervised Learning for Semantic Multi-Style Transfer

Abstract

Numerous methods for style transfer have been developed using unsupervised learning and gained impressive results. However, optimal style transfer cannot be conducted from a global fashion in certain style domains, mainly when a single target-style domain contains semantic objects that have their own distinct and unique styles, e.g., those objects in the anime-style domain. Previous methods are incongruent because the unsupervised learning can not provide the semantic mappings between the multistyle objects according to their unique styles. Thus, in this paper, we propose a pseudo-supervised learning framework for the semantic multi-style transfer (SMST), which consists of (i) a pseudo ground truth (pGT) generation phase and (ii) a SMST learning phase. In the pGT generation phase, multiple semantic objects of the photo images are separately transferred to the target-domain object styles in an object-oriented fashion. Then the transferred objects are composed back to an image, which is the pGT. In the SMST learning phase, a SMST network (SMSTnet) is trained with the pairs of the photo images and its respective pGT in a supervised manner. From this, our framework can provide the semantic mappings of multi-style objects. Moreover, to embrace the multi-styles of various objects into a single generator, we design the SMSTnet with channel attentions in conjunction with a discriminator dedicated to our pseudo-supervised learning. Our method has been applied and intensively tested for anime-style transfer learning. The experimental results demonstrate the effectiveness of our method and show its superiority compared to the state-of-theart methods.

Application of in Information Visualization and Generative Adversarial Network in the Review of Style Transfer

“…In the other words, when they have their own distinct and unique styles, such as anime-style domain, the current style transfer is not working. Therefore [11] propose pseudosupervised learning framework, it provides the semantic connections between the objects of the two style domain in the image space. They introduce a generator network SMSTnet and a new training method for the discriminator.…”

Dress-up: Deep Neural Framework for Image-based Human Appearance Transfer

“…GANs have achieved a high-grade visual quality in different image processing problems such as conditional image synthesis [18][19][20][21][22] and style transfer [23][24][25]. In our work, we have interested to the conditional image synthesis to generate images conditioning on specific constraint task.…”