UltraPixel：将超高分辨率图像合成推向新高峰
UltraPixel: Advancing Ultra-High-Resolution Image Synthesis to New Peaks
AI总结：
UltraPixel 提出一种级联扩散模型架构，利用低分辨率语义引导、隐式神经表示和尺度感知归一化，在单一模型中高效生成 1K 至 6K 超高分辨率图像，仅增加不到 3% 参数即达到最先进性能。
摘要：
超高分辨率图像生成面临巨大挑战，例如语义规划复杂度增加、细节合成困难，以及巨大的训练资源需求。我们提出 UltraPixel，这是一种利用级联扩散模型在单一模型内生成多种分辨率（例如 1K 到 6K）高质量图像的新颖架构，同时保持计算效率。UltraPixel 在后期去噪阶段利用低分辨率图像中语义丰富的表示来引导整个高细节高分辨率图像的生成，显著降低了复杂度。此外，我们引入了用于连续上采样的隐式神经表示和可适配各种分辨率的尺度感知归一化层。值得注意的是，低分辨率和高分辨率过程都在最紧凑的空间中进行，共享大部分参数，高分辨率输出仅增加不到 3% 的额外参数，大幅提升了训练和推理效率。我们的模型实现了快速训练并降低了数据需求，能够生成照片级真实感的高分辨率图像，并在大量实验中展现出最先进的性能。
英文摘要：
Ultra-high-resolution image generation poses great challenges, such as increased semantic planning complexity and detail synthesis difficulties, alongside substantial training resource demands. We present UltraPixel, a novel architecture utilizing cascade diffusion models to generate high-quality images at multiple resolutions (\textit{e.g.}, 1K to 6K) within a single model, while maintaining computational efficiency. UltraPixel leverages semantics-rich representations of lower-resolution images in the later denoising stage to guide the whole generation of highly detailed high-resolution images, significantly reducing complexity. Furthermore, we introduce implicit neural representations for continuous upsampling and scale-aware normalization layers adaptable to various resolutions. Notably, both low- and high-resolution processes are performed in the most compact space, sharing the majority of parameters with less than 3$\%$ additional parameters for high-resolution outputs, largely enhancing training and inference efficiency. Our model achieves fast training with reduced data requirements, producing photo-realistic high-resolution images and demonstrating state-of-the-art performance in extensive experiments.
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。
