Navigating the GAN Parameter Space for Semantic Image Editing
By Anton Cherepkov, Andrey Voynov, Artem Babenko. 2020
79 citations recorded by the source corpus.
Generative Adversarial Networks (GANs) are currently an indispensable tool for visual editing, being a standard component of image-to-image translation and image restoration pipelines. Furthermore, GANs are especially advantageous for controllable generation since their latent spaces contain a wide range of interpretable directions, well suited for semantic editing operations. By gradually changing latent codes along these directions, one can produce impressive visual effects, unattainable without GANs.In this paper, we significantly expand the range of visual effects achievable with the state-of-the-art models, like StyleGAN2. In contrast to existing works, which mostly operate by latent codes, we discover interpretable directions in the space of the generator parameters. By several simple methods, we explore this space and demonstrate that it also contains a plethora of interpretable directions, which are an excellent source of non-trivial semantic manipulations. The discovered manipulations cannot be achieved by transforming the latent codes and can be used to edit both synthetic and real images. We release our code and models and hope they will serve as a handy tool for further efforts on GAN-based image editing.
Subfields: Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Dacheng Tao
- Sanja Fidler
- Antonio Torralba
- Ming-Hsuan Yang
- Jitendra Malik
- Trevor Darrell
- Xiaogang Wang
Citations in the corpus
- EditGAN: High-Precision Semantic Image Editing (2021, methodology): “[16] proposed an approach that more directly analyses the parameter space of a GAN and treats it as a latent space in which to apply edits.”
- SVDiff: Compact Parameter Space for Diffusion Fine-Tuning (2023, methodology): “Instead of fine-tuning the full weight matrix, we only up-date the weight matrix by optimizing the spectral shift [13], δ , which is defined as the difference between the singular values of the updated weight matrix and the original weight matrix.”