研究人员开发了一种从单张图像生成完整3D场景网格的新方法，将现有的以对象为中心的生成器（如Trellis.2）扩展到处理室内外环境。该方法根据距离将场景划分为自适应块，捕获显式的2D-3D对应关系，并合成新的室外场景数据以克服现有数据集的局限性。在基准数据集和真实图像上的实验表明，与以前的方法相比，该方法具有更高的几何精度和感知质量。 AI
影响 这项进展可能导致从有限输入中更高效、更详细地创建3D内容，影响虚拟现实和自主系统等领域。
排序理由 该集群描述了一篇详细介绍新颖3D场景生成方法的研究论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
Single-image scene generation aims to produce a complete 3D scene mesh from a single image, including surfaces the camera did not observe. While pretrained 3D object generators encode a strong shape prior, they are mainly designed for isolated objects in a fixed canonical volume …
Single-image scene generation aims to produce a complete 3D scene mesh from a single image, including surfaces the camera did not observe. While pretrained 3D object generators encode a strong shape prior, they are mainly designed for isolated objects in a fixed canonical volume …
arXiv:2610.08790v1 Announce Type: new Abstract: Single-image scene generation aims to produce a complete 3D scene mesh from a single image, including surfaces the camera did not observe. While pretrained 3D object generators encode a strong shape prior, they are mainly designed f…