Karras, TeroLaine, SamuliLaine, SamuliLehtinen, JaakkoAila, Timo2025-06-022025-06-022018-12-05http://data.inu.ac.kr/handle/123456789/1938StyleGAN은 고해상도 얼굴 이미지 생성을 위한 스타일 기반 GAN 구조를 제시하여 디테일 제어와 잠재 공간 분리성에서 우수한 성능을 보여줍니다 ©2018 NVIDIAWe propose an alternative generator architecture for generative adversarial networks, borrowing from style transfer literature. The new architecture leads to an automatically learned, unsupervised separation of high-level attributes (e.g., pose and identity when trained on human faces) and stochastic variation in the generated images (e.g., freckles, hair), and it enables intuitive, scale-specific control of the synthesis. The new generator improves the state-of-the-art in terms of traditional distribution quality metrics, leads to demonstrably better interpolation properties, and also better disentangles the latent factors of variation. To quantify interpolation quality and disentanglement, we propose two new, automated methods that are applicable to any generator architecture. Finally, we introduce a new, highly varied and high-quality dataset of human faces.en-USStyleGANGANImage GenerationComputer VisionNeural NetworksArticle