研究人员开发了 GenRe，一个扩散引导式系统，用于增强自动驾驶模拟的城市场景重建。该方法通过学习各种场景的生成先验，提高了 3D 表示的质量，尤其是在具有挑战性的视角下。GenRe 能在几分钟内高效修复现有 3D 高斯表示的缺陷，提供稳健且高保真的结果，这些结果可以泛化到未见的视角，并有利于传感器模拟等下游任务。 AI
影响 通过增强具有挑战性视角下的 3D 场景重建质量，改进了自动驾驶的传感器模拟。
排序理由 该集群包含一篇详细介绍城市场景重建新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
Urban scene reconstruction from real-world observations has emerged as a powerful tool for self-driving development and testing. While current neural rendering approaches achieve high-fidelity rendering along the recorded trajectories, their quality degrades significantly under l…
arXiv:2605.22420v1 Announce Type: new Abstract: Urban scene reconstruction from real-world observations has emerged as a powerful tool for self-driving development and testing. While current neural rendering approaches achieve high-fidelity rendering along the recorded trajectori…