arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2504.13622eess.IVcs.CV
收藏

SupResDiffGAN:一种用于超分辨率任务的新方法

SupResDiffGAN a new approach for the Super-Resolution task

WUST, Wybrzeże Stanisława Wyspiańskiego 27, 50-370 Wrocław, Poland(波兰弗罗茨瓦夫大学)

Dawid Kopeć, Wojciech Kozłowski, Maciej Wizerkaniuk, Dawid Krutul, Jan Kocoń, Maciej Zięba

AI总结:

该研究提出混合架构SupResDiffGAN,结合GAN与扩散模型优势,通过潜在空间表示和减少扩散步骤实现更快推理,用自适应噪声破坏防判别器过拟合,实验优于SR3等模型,弥合两类方法性能差距。

摘要:

在这项工作中,我们提出了SupResDiffGAN,这是一种新颖的混合架构,它结合了生成对抗网络(Generative Adversarial Networks,GANs)和扩散模型在超分辨率任务中的优势。通过利用潜在空间表示并减少扩散步骤的数量,SupResDiffGAN实现了比其他基于扩散的超分辨率模型显著更快的推理时间,同时保持了具有竞争力的感知质量。为了防止判别器过拟合,我们提出了自适应噪声破坏,确保训练期间生成器和判别器之间的稳定平衡。在基准数据集上的大量实验表明,我们的方法在效率和图像质量上优于SR3和I²SB等传统扩散模型。这项工作弥合了基于扩散和基于GAN的方法之间的性能差距,为扩散模型在高分辨率图像生成中的实时应用奠定了基础。

英文摘要:

In this work, we present SupResDiffGAN, a novel hybrid architecture that combines the strengths of Generative Adversarial Networks (GANs) and diffusion models for super-resolution tasks. By leveraging latent space representations and reducing the number of diffusion steps, SupResDiffGAN achieves significantly faster inference times than other diffusion-based super-resolution models while maintaining competitive perceptual quality. To prevent discriminator overfitting, we propose adaptive noise corruption, ensuring a stable balance between the generator and the discriminator during training. Extensive experiments on benchmark datasets show that our approach outperforms traditional diffusion models such as SR3 and I$^2$SB in efficiency and image quality. This work bridges the performance gap between diffusion- and GAN-based methods, laying the foundation for real-time applications of diffusion models in high-resolution image generation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑