研究人员开发了SHERPA，一个旨在适应大规模文本到图像模型以生成360度全景图的新框架。现有模型在等距柱状投影（ERP）全景图的独特拓扑结构方面存在困难，导致错位，尤其是在接缝和极地区域。SHERPA通过引入频率选择RoPE、圆形编码和双路径训练方案来解决这一问题，从而能够生成写实和风格化的全景场景。 AI
影响 使文本到图像模型能够生成更准确和风格化的360度全景图。
排序理由 该集群包含一篇详细介绍适应现有模型的新框架的研究论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2606.12213v1 Announce Type: new Abstract: Panoramic imagery is increasingly used in world-generation, games, and simulation, where users may need not only photorealistic scenes but also stylized and non-photorealistic environments. Large-scale text-to-image diffusion and fl…
Panoramic imagery is increasingly used in world-generation, games, and simulation, where users may need not only photorealistic scenes but also stylized and non-photorealistic environments. Large-scale text-to-image diffusion and flow models provide broad style and semantic prior…