Lower dimensional kernels for video discriminators

Emmanuel Kahembwe, Subramanian Ramamoorthy

Research output: Contribution to journalArticle

1 Downloads (Pure)

Abstract

This work presents an analysis of the discriminators used in Generative Adversarial Networks (GANs) for Video. We show that unconstrained video discriminator architectures induce a loss surface with high curvature which make optimization difficult. We also show that this curvature becomes more extreme as the maximal kernel dimension of video discriminators increases. With these observations in hand, we propose a methodology for the design of a family of efficient Lower-Dimensional Video Discriminators for GANs (LDVD-GANs). The proposed methodology improves the performance and efficiency of video GAN models it is applied to and demonstrates good performance on complex and diverse datasets such as UCF-101. In particular, we show that LDVDs can double the performance of Temporal-GANs and provide for state-of-the-art performance on a single GPU using the proposed methodology.

Original languageEnglish
Pages (from-to)506-520
Number of pages15
JournalNeural Networks
Volume132
Early online date26 Sep 2020
DOIs
Publication statusE-pub ahead of print - 26 Sep 2020

Keywords

  • Discriminator analysis
  • Generative Adversarial Networks
  • Video generation

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Lower dimensional kernels for video discriminators'. Together they form a unique fingerprint.

  • Cite this