Unsupervised Discovery of Disentangled Manifolds in GANs

Yu-Ding Lu, Hsin-Ying Lee, Hung-Yu Tseng, Ming-Hsuan Yang

As recent generative models can generate photo-realistic images, people seek to understand the mechanism behind the generation process. Interpretable generation process is beneficial to various image editing applications. In this work, we propose a framework to discover interpretable directions in the latent space given arbitrary pre-trained generative adversarial networks. We propose to learn the transformation from prior one-hot vectors representing different attributes to the latent space used by pre-trained models. Furthermore, we apply a centroid loss function to improve consistency and smoothness while traversing through different directions. We demonstrate the efficacy of the proposed framework on a wide range of datasets. The discovered direction vectors are shown to be visually corresponding to various distinct attributes and thus enable attribute editing.

Knowledge Graph

arrow_drop_up

Comments

Sign up or login to leave a comment