Help Privacy Policy Disclaimer
  Advanced SearchBrowse





Discovering Class-Specific GAN Controls for Semantic Image Synthesis


Schiele,  Bernt
Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society;

External Resource
No external resources are shared
Fulltext (restricted access)
There are currently no full texts shared for your IP range.
Fulltext (public)

(Preprint), 11MB

Supplementary Material (public)
There is no public supplementary material available

Schönfeld, E., Borges, J., Sushko, V., Schiele, B., & Khoreva, A. (2022). Discovering Class-Specific GAN Controls for Semantic Image Synthesis. Retrieved from https://arxiv.org/abs/2212.01455.

Cite as: https://hdl.handle.net/21.11116/0000-000C-185D-2
Prior work has extensively studied the latent space structure of GANs for
unconditional image synthesis, enabling global editing of generated images by
the unsupervised discovery of interpretable latent directions. However, the
discovery of latent directions for conditional GANs for semantic image
synthesis (SIS) has remained unexplored. In this work, we specifically focus on
addressing this gap. We propose a novel optimization method for finding
spatially disentangled class-specific directions in the latent space of
pretrained SIS models. We show that the latent directions found by our method
can effectively control the local appearance of semantic classes, e.g.,
changing their internal structure, texture or color independently from each
other. Visual inspection and quantitative evaluation of the discovered GAN
controls on various datasets demonstrate that our method discovers a diverse
set of unique and semantically meaningful latent directions for class-specific