Trained Latent Space Navigation to Prevent Lack of Photorealism in Generated Images on Style-based Models
This work addresses a specific issue in style-based generative models for researchers and practitioners, offering an incremental improvement to enhance image quality.
The paper tackles the problem of generated images lacking photorealism in StyleGAN models by proposing an unsupervised method that identifies and restricts latent code manipulations to well-trained local subspaces, resulting in maintained photorealism even with significant manipulations.
Recent studies on StyleGAN variants show promising performances for various generation tasks. In these models, latent codes have traditionally been manipulated and searched for the desired images. However, this approach sometimes suffers from a lack of photorealism in generated images due to a lack of knowledge about the geometry of the trained latent space. In this paper, we show a simple unsupervised method that provides well-trained local latent subspace, enabling latent code navigation while preserving the photorealism of the generated images. Specifically, the method identifies densely mapped latent spaces and restricts latent manipulations within the local latent subspace. Experimental results demonstrate that images generated within the local latent subspace maintain photorealism even when the latent codes are significantly and repeatedly manipulated. Moreover, experiments show that the method can be applied to latent code optimization for various types of style-based models. Our empirical evidence of the method will benefit applications in style-based models.