Delta-GAN-Encoder: Encoding Semantic Changes for Explicit Image Editing, using Few Synthetic Samples
This addresses the challenge of precise and continuous image editing without structural constraints on latent spaces, which is incremental but useful for applications in facial imagery.
The paper tackles the problem of controlling attributes in a pre-trained GAN's latent space for image editing by proposing a method that learns to encode semantic changes using few synthetic samples, achieving state-of-the-art results in facial imagery editing.
Understating and controlling generative models' latent space is a complex task. In this paper, we propose a novel method for learning to control any desired attribute in a pre-trained GAN's latent space, for the purpose of editing synthesized and real-world data samples accordingly. We perform Sim2Real learning, relying on minimal samples to achieve an unlimited amount of continuous precise edits. We present an Autoencoder-based model that learns to encode the semantics of changes between images as a basis for editing new samples later on, achieving precise desired results - example shown in Fig. 1. While previous editing methods rely on a known structure of latent spaces (e.g., linearity of some semantics in StyleGAN), our method inherently does not require any structural constraints. We demonstrate our method in the domain of facial imagery: editing different expressions, poses, and lighting attributes, achieving state-of-the-art results.