https://gentleostrich.github.io/categories/%E4%B8%89%E7%BB%B4%E9%87%8D%E5%BB%BA/Recent content in 三维重建 on GentleOstrichHugozh-HansThu, 15 Jan 2026 00:00:00 +0000https://gentleostrich.github.io/posts/sam-3d-object-%E6%95%99%E7%A8%8B/Thu, 15 Jan 2026 00:00:00 +0000https://gentleostrich.github.io/posts/sam-3d-object-%E6%95%99%E7%A8%8B/<h2 id="安装" class="relative group">安装 <span class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100"><a class="group-hover:text-primary-300 dark:group-hover:text-neutral-700" style="text-decoration-line: none !important;" href="#%e5%ae%89%e8%a3%85" aria-label="锚点">#</a></span></h2><p>参考<a href="https://github.com/facebookresearch/sam-3d-objects/blob/main/doc/setup.md" target="_blank" rel="noreferrer">官方安装教程</a>在服务器上配置环境，服务器全程使用 clash 代理，使用 conda 安装</p> <h3 id="环境配置" class="relative group">环境配置 <span class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100"><a class="group-hover:text-primary-300 dark:group-hover:text-neutral-700" style="text-decoration-line: none !important;" href="#%e7%8e%af%e5%a2%83%e9%85%8d%e7%bd%ae" aria-label="锚点">#</a></span></h3><div class="highlight"><pre tabindex="0" class="chroma"><code class="language-shell" data-lang="shell"><span class="line"><span class="cl"><span class="c1"># 克隆项目仓库</span> </span></span><span class="line"><span class="cl">git clone https://github.com/facebookresearch/sam-3d-objects.git </span></span><span class="line"><span class="cl"><span class="nb">cd</span> sam-3d-objects </span></span><span class="line"><span class="cl"> </span></span><span class="line"><span class="cl"><span class="c1"># 创建 conda 环境</span> </span></span><span class="line"><span class="cl">conda env create -f environments/default.yml </span></span><span class="line"><span class="cl">conda activate sam3d-objects </span></span><span class="line"><span class="cl"> </span></span><span class="line"><span class="cl"><span class="c1"># 设置 pytorch 等安装的 URL</span> </span></span><span class="line"><span class="cl"><span class="nb">export</span> <span class="nv">PIP_EXTRA_INDEX_URL</span><span class="o">=</span><span class="s2">&#34;https://pypi.ngc.nvidia.com https://download.pytorch.org/whl/cu121&#34;</span> </span></span><span class="line"><span class="cl"> </span></span><span class="line"><span class="cl"><span class="c1"># 安装 sam 3d 核心依赖</span> </span></span><span class="line"><span class="cl">pip install -e <span class="s1">&#39;.[dev]&#39;</span> </span></span><span class="line"><span class="cl">pip install -e <span class="s1">&#39;.[p3d]&#39;</span> <span class="c1"># pytorch3d dependency on pytorch is broken, this 2-step approach solves it</span> </span></span><span class="line"><span class="cl"> </span></span><span class="line"><span class="cl"><span class="c1"># 安装 sam 3d 推理</span> </span></span><span class="line"><span class="cl"><span class="nb">export</span> <span class="nv">PIP_FIND_LINKS</span><span class="o">=</span><span class="s2">&#34;https://nvidia-kaolin.s3.us-east-2.amazonaws.com/torch-2.5.1_cu121.html&#34;</span> </span></span><span class="line"><span class="cl">pip install -e <span class="s1">&#39;.[inference]&#39;</span> </span></span></code></pre></div><h3 id="下载官方权重文件" class="relative group">下载官方权重文件 <span class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100"><a class="group-hover:text-primary-300 dark:group-hover:text-neutral-700" style="text-decoration-line: none !important;" href="#%e4%b8%8b%e8%bd%bd%e5%ae%98%e6%96%b9%e6%9d%83%e9%87%8d%e6%96%87%e4%bb%b6" aria-label="锚点">#</a></span></h3><p>首先需要在 huggingface 上申请许可，笔者使用新邮箱创建了一个新的 huggingface 账户，在申请许可需要填入的个人信息中填写美国地址、学校等等，否则容易被拒。获得许可后，执行：</p>