stable diffusion
Ciência da Computação
Estatística
Verbete 1 de 1
substantivo
Contexto: "The participants’ comments about art generated AI brings us over to new questions: Please look at the picture in Fig.4. It is created by AI. This tiny experiment was conducted in November 2022 with the help of our colleague in his office. The algorithm creates a picture based on existing pictures online and uses the tagging that people provide. The model, or concept is called stable diffusion and is an example of deep learning as depicted in one of the branches in Fig.1 above."
Fonte: Presthus, W., & Sørum, H. (2025). Consumer Perspectives on the Increasing Use of Artificial Intelligence: It is Helpful, Scary or So Wrong? Procedia Computer Science, 256, 158–165. https://doi.org/10.1016/j.procs.2025.02.108
Fonte: Presthus, W., & Sørum, H. (2025). Consumer Perspectives on the Increasing Use of Artificial Intelligence: It is Helpful, Scary or So Wrong? Procedia Computer Science, 256, 158–165. https://doi.org/10.1016/j.procs.2025.02.108
Termo equivalente:
Definição: "Stable diffusion is a latent text-to-image diffusion model (DM) which serves as the foundation for an array of recent advancements in, e.g., 3D classification, controllable image editing, image personalization, synthetic data augmentation, graphical user interface prototyping, etc."
Fonte: Podell, D., English, Z., Lacey, K., Blattmann, A., Dockhorn, T., Müller, J., Penna, J., & Rombach, R. (2023). SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis. arXiv. https://doi.org/10.48550/arXiv.2307.01952
Fonte: Podell, D., English, Z., Lacey, K., Blattmann, A., Dockhorn, T., Müller, J., Penna, J., & Rombach, R. (2023). SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis. arXiv. https://doi.org/10.48550/arXiv.2307.01952
Definição em português: "Difusão estável é um modelo de difusão de texto para imagem que serve como base para uma série de avanços recentes em, por exemplo, classificação 3D, edição de imagem controlável, personalização de imagem, aumento de dados sintéticos, prototipagem de interface gráfica de usuário, etc."