Cluster with GANs

Yuri Feigin*, Hedva Spitzer, Raja Giryes

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Generative adversarial networks (GANs) and clustering algorithms are both very popular unsupervised methodologies in machine learning. In this work, we propose a novel strategy that uses GANs to improve clustering and vice verse. We start by providing a simple but yet powerful scheme for improving clustering methods that rely on the latent space of GANs. Then, we turn to demonstrate how the output of clustering techniques can be employed for significantly improving the output quality of GANs. We empirically demonstrate the improvement that is achieved by our proposed framework both for clustering and the generation quality of GANs measured by the inception score.

Original languageEnglish
Article number103571
JournalComputer Vision and Image Understanding
Volume225
DOIs
StatePublished - Dec 2022

Keywords

  • clustering
  • GAN

Fingerprint

Dive into the research topics of 'Cluster with GANs'. Together they form a unique fingerprint.

Cite this