1. Linear Semantics in Generative Adversarial Networks
- Author
-
Jianjin Xu and Changxi Zheng
- Subjects
FOS: Computer and information sciences ,Computer science ,Computer Science - Artificial Intelligence ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition ,02 engineering and technology ,Image editing ,computer.software_genre ,Semantics ,Facial recognition system ,030218 nuclear medicine & medical imaging ,Image (mathematics) ,03 medical and health sciences ,0302 clinical medicine ,0202 electrical engineering, electronic engineering, information engineering ,Code (cryptography) ,Segmentation ,business.industry ,Image segmentation ,Feature (linguistics) ,Artificial Intelligence (cs.AI) ,020201 artificial intelligence & image processing ,Artificial intelligence ,business ,computer ,Natural language processing - Abstract
Generative Adversarial Networks (GANs) are able to generate high-quality images, but it remains difficult to explicitly specify the semantics of synthesized images. In this work, we aim to better understand the semantic representation of GANs, and thereby enable semantic control in GAN’s generation process. Interestingly, we find that a well-trained GAN encodes image semantics in its internal feature maps in a surprisingly simple way: a linear transformation of feature maps suffices to extract the generated image semantics. To verify this simplicity, we conduct extensive experiments on various GANs and datasets; and thanks to this simplicity, we are able to learn a semantic segmentation model for a trained GAN from a small number (e.g., 8) of labeled images. Last but not least, leveraging our finding, we propose two few-shot image editing approaches, namely Semantic-Conditional Sampling and Semantic Image Editing. Given a trained GAN and as few as eight semantic annotations, the user is able to generate diverse images subject to a user-provided semantic layout, and control the synthesized image semantics. We have made the code publicly available1.
- Published
- 2021
- Full Text
- View/download PDF