Latent Transformations via NeuralODEs for GAN-based Image Editing
This work addresses image editing challenges for domains beyond faces, offering a method to handle more complex variations, though it is incremental as it builds on existing GAN-based approaches.
The paper tackles the problem of semantic image editing in non-face domains by showing that nonlinear latent transformations via Neural ODEs outperform linear shifts for complex attributes, demonstrating improved controllability on datasets with known attributes.
Recent advances in high-fidelity semantic image editing heavily rely on the presumably disentangled latent spaces of the state-of-the-art generative models, such as StyleGAN. Specifically, recent works show that it is possible to achieve decent controllability of attributes in face images via linear shifts along with latent directions. Several recent methods address the discovery of such directions, implicitly assuming that the state-of-the-art GANs learn the latent spaces with inherently linearly separable attribute distributions and semantic vector arithmetic properties. In our work, we show that nonlinear latent code manipulations realized as flows of a trainable Neural ODE are beneficial for many practical non-face image domains with more complex non-textured factors of variation. In particular, we investigate a large number of datasets with known attributes and demonstrate that certain attribute manipulations are challenging to obtain with linear shifts only.