<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[KoMiRa的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/KoMiRa</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; KoMiRa]]></copyright><item><title><![CDATA[（自用）pointnet++&RANSAN&ICP进行点云配准]]></title><link>https://blog.youkuaiyun.com/KoMiRa/article/details/158658042</link><guid>https://blog.youkuaiyun.com/KoMiRa/article/details/158658042</guid><author>KoMiRa</author><pubDate>Wed, 04 Mar 2026 17:00:58 +0800</pubDate><description><![CDATA[本文探讨了点云配准中ICP算法的局限性及其改进方法。针对ICP算法对初始位置敏感的问题，提出结合RANSAC粗配准和PointNet++深度特征提取的解决方案。实验采用StanfordBunny数据集，通过RANSAC获得初始变换后，再使用ICP进行精配准。结果表明，该方法能有效避免局部最优问题，实现点云的高精度配准。与传统特征提取方法相比，基于PointNet++的深度特征提取展现出更优性能。配准评估指标显示误差接近于零，验证了该方法的有效性。]]></description><category></category></item><item><title><![CDATA[python简单程序]]></title><link>https://blog.youkuaiyun.com/KoMiRa/article/details/125399093</link><guid>https://blog.youkuaiyun.com/KoMiRa/article/details/125399093</guid><author>KoMiRa</author><pubDate>Tue, 21 Jun 2022 21:46:24 +0800</pubDate><description><![CDATA[python简单程序]]></description><category></category></item></channel></rss>