研究人员开发了 JanusMesh，一个用于生成文本驱动的 3D 视觉错觉的新颖框架。该方法采用两阶段过程：首先，跨空间双分支去噪技术对齐 3D 潜在表示并融合符号距离场以实现几何一致性；其次，视图条件纹理合成模块整合 2D 扩散先验。JanusMesh 可以在 3-5 分钟内生成逼真的双语义 3D 错觉，在效率和质量方面优于现有方法。 AI
影响 能够为各种应用更快、更有效地创建复杂的 3D 视觉错觉。
排序理由 该集群包含一篇详细介绍 3D 视觉错觉生成新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
A fast, training-free framework generates text-driven 3D visual illusions by decoupling generation into cross-space dual-branch denoising and view-conditioned texture synthesis for seamless geometric fusion and semantic coherence.
arXiv:2606.20563v1 Announce Type: new Abstract: Creating 3D visual illusions, a single 3D mesh that reveals entirely different semantics from various viewing angles, is a fascinating but tough challenge. Existing optimization-based methods are slow and can produce oversaturated c…
Creating 3D visual illusions, a single 3D mesh that reveals entirely different semantics from various viewing angles, is a fascinating but tough challenge. Existing optimization-based methods are slow and can produce oversaturated colors. In contrast, naive stitching approaches f…