<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>图像分割 on Text Matrix</title><link>https://txtmix.com/tags/%E5%9B%BE%E5%83%8F%E5%88%86%E5%89%B2/</link><description>Recent content in 图像分割 on Text Matrix</description><generator>Hugo</generator><language>zh-cn</language><lastBuildDate>Tue, 21 Jul 2026 20:06:14 +0800</lastBuildDate><atom:link href="https://txtmix.com/tags/%E5%9B%BE%E5%83%8F%E5%88%86%E5%89%B2/index.xml" rel="self" type="application/rss+xml"/><item><title>SAM 3：Meta 开源第三代分割一切模型</title><link>https://txtmix.com/posts/tech/sam3-segment-anything-model-3/</link><pubDate>Sat, 23 May 2026 13:09:23 +0800</pubDate><guid>https://txtmix.com/posts/tech/sam3-segment-anything-model-3/</guid><description>&lt;h1 id="sam-3-segment-anything-model-3-meta-开源的第三代分割一切模型">SAM 3 (Segment Anything Model 3): Meta 开源的第三代分割一切模型&lt;/h1>
&lt;p>&lt;strong>分类：&lt;/strong> CV · 图像分割
&lt;strong>Stars：&lt;/strong> 10,061
&lt;strong>地址：&lt;/strong> &lt;a href="https://github.com/facebookresearch/sam3" target="_blank" rel="noopener noreffer ">https://github.com/facebookresearch/sam3&lt;/a>
&lt;strong>官网：&lt;/strong> &lt;a href="https://ai.meta.com/sam/" target="_blank" rel="noopener noreffer ">https://ai.meta.com/sam/&lt;/a>&lt;/p>
&lt;p>SAM 3 不只是 SAM 2 的升级版。它真正的分水岭在于：从“在 2D 图像里把东西圈出来”跨到了“在 3D 世界里理解物体的空间边界”。点云、mesh、NeRF 场景都可以直接丢进去做分割，而且模型更小、跑得更快。&lt;/p></description></item></channel></rss>