EE Seminar: Unsupervised K-modal Styled Content Generation

17 בפברואר 2020, 15:00 
Room 012, Kitot Building 

Speaker: Omry Sendik

Ph.D. student under the supervision of Prof. Daniel Cohen-Or and Prof. Shai Avidan

 

Monday, February 17th, 2020 at 15:00
Room 206, Wolfson Mechanical Eng. Bldg., Faculty of Engineering

Unsupervised K-modal Styled Content Generation

 

Abstract

 

The emergence of deep generative models has recently enabled the automatic generation of massive amounts of graphical content, both in 2D and in 3D.
Generative Adversarial Networks (GANs) and style control mechanisms, such as Adaptive Instance Normalization (AdaIN), have proved particularly effective in this context, culminating in the state-of-the-art StyleGAN architecture.
While such models are able to learn diverse distributions, provided a sufficiently large training set, they are not well-suited for scenarios where the distribution of the training data exhibits a multi-modal behavior. In such cases, reshaping a uniform or normal distribution over the latent space into a complex multi-modal distribution in the data domain is challenging, and the generator might fail to sample the target distribution well. Furthermore, existing unsupervised generative models are not able to control the mode of the generated samples independently of the other visual attributes, despite the fact that they are typically disentangled in the training data.  

In this work, we introduce uMM-GAN, a novel architecture designed to better model multi-modal distributions, in an unsupervised fashion.
Building upon the StyleGAN architecture, our network learns multiple modes, in a completely unsupervised manner, and combines them using a set of learned weights.
We demonstrate that this approach is capable of effectively approximating a complex distribution as a superposition of multiple simple ones.
We further show that uMM-GAN effectively disentangles between modes and style, thereby providing an independent degree of control over the generated content.

אוניברסיטת תל אביב עושה כל מאמץ לכבד זכויות יוצרים. אם בבעלותך זכויות יוצרים בתכנים שנמצאים פה ו/או השימוש שנעשה בתכנים אלה לדעתך מפר זכויות, נא לפנות בהקדם לכתובת שכאן >>
אוניברסיטת תל-אביב, ת.ד. 39040, תל-אביב 6997801
UI/UX Basch_Interactive