Do Not Escape From the Manifold: Discovering the Local Coordinates on the Latent Space of GANs
This addresses the challenge of limited success in global traversal directions for GANs, particularly in StyleGAN2's W-space, which is incremental as it builds on prior disentanglement research.
The paper tackles the problem of finding semantically meaningful directions in GAN latent spaces by proposing an unsupervised method based on local geometry, showing that local principal variations correspond to semantic factorization and provide robust image traversal.
The discovery of the disentanglement properties of the latent space in GANs motivated a lot of research to find the semantically meaningful directions on it. In this paper, we suggest that the disentanglement property is closely related to the geometry of the latent space. In this regard, we propose an unsupervised method for finding the semantic-factorizing directions on the intermediate latent space of GANs based on the local geometry. Intuitively, our proposed method, called Local Basis, finds the principal variation of the latent space in the neighborhood of the base latent variable. Experimental results show that the local principal variation corresponds to the semantic factorization and traversing along it provides strong robustness to image traversal. Moreover, we suggest an explanation for the limited success in finding the global traversal directions in the latent space, especially W-space of StyleGAN2. We show that W-space is warped globally by comparing the local geometry, discovered from Local Basis, through the metric on Grassmannian Manifold. The global warpage implies that the latent space is not well-aligned globally and therefore the global traversal directions are bound to show limited success on it.