# 并行推理

Source: [https://docs.qualcomm.com/doc/80-70018-50SC/topic/gst-ai-parallel-inference.html](https://docs.qualcomm.com/doc/80-70018-50SC/topic/gst-ai-parallel-inference.html)

**gst-ai-parallel-inference** 应用程序可以对来自不同源（例如摄像头、文件或 RTSP 网络）的输入流执行目标检测、对象分类、姿态检测和图像分割。这些用例使用 LiteRT 模型进行目标检测、图像分割、分类和姿态检测。

Note: 此应用在 QCS8275 上不支持。

该图显示了一个 pipeline，该 pipeline 从摄像头、文件或 RTSP 流获取输入，对四个用例执行并行推理，并在屏幕上并排显示结果。

有关 pipeline 流中使用的插件的信息，请参见 [Pipeline 流](https://docs.qualcomm.com/doc/80-70018-50SC/topic/gst-ai-parallel-inference.html#gst-ai-parallel-inference__section_gcg_r3s_lbc)。

Figure : gst-ai-parallel-inference pipeline
            
            ![](data:image/png;base64,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)

通过在一个实时摄像头流上执行四个并行 AI 推理来增强您的开发项目。该视频展示了如何设置 pipeline 并且并排显示结果的分步过程。视频将在新页面中打开。

<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewbox="0 0 640 400" width="640" height="400" style="cursor:auto !important">
    <defs>
      <style>@import url("https://fonts.googleapis.com/css2?family=Roboto+Flex:opsz,wght@8..144,100..1000&amp;display=swap");
.svg-1 .bg-fill { fill: var(--color-background) }
.svg-1 .fill-text { color: var(--color-content); fill: var(--color-content) }
.svg-1 .video-hoverbox { transition: opacity 0.15s ease-in-out }
.svg-1 .video-hoverbox:hover { opacity: 0.9 }</style>
  </defs>

  <foreignobject x="0" y="0" width="640" height="400">
    <body xmlns="http://www.w3.org/1999/xhtml">
        <iframe width="640" height="400" src="https://players.brightcove.net/1414329538001/BJv5wEFt_default/index.html?videoId=6355769124112" allowfullscreen="" allow="encrypted-media"></iframe>
    <div class='topic-detail'><div class='topic-updated-date'><span> Last Published: </span>Nov 12, 2025</div><div class='prev-and-next-links'><span class='previous-topic-link'><span aria-hidden='true' class='disabled' data-tip='' data-effect='solid'></span></span></div></div></body>
    </foreignobject>
</svg>

## 示例模型和标签文件

Table : gst-ai-parallel-inference 的示例模型和标签文件

| 应用程序 | 模型文件 | 标签文件 |
| --- | --- | --- |
| 目标检测 | <var class="keyword varname">yolov8_det_quantized.tflite</var> | <var class="keyword varname">yolov8.labels</var> |
| 姿态估计 | <var class="keyword varname">hrnet_pose_quantized.tflite </var> | <var class="keyword varname">hrnet_pose.labels</var> |
| 分割 | <var class="keyword varname">deeplabv3_plus_mobilenet_quantized.tflite<br>                                </var> | <var class="keyword varname">deeplabv3_resnet50.labels </var> |
| 分类 | <var class="keyword varname">inception_v3_quantized.tflite</var> | <var class="keyword varname">classification.labels </var> |

## 前提条件

- 如果还没有安装 eSDK，请[下载并安装 eSDK](https://docs.qualcomm.com/bundle/publicresource/topics/80-70018-51/install-sdk.html#download-and-install-esdk-)。
- [下载模型和标签文件](https://docs.qualcomm.com/doc/80-70018-50SC/topic/download-model-and-label-files.html)。
- 要访问主机，请启用 SSH。有关说明，请参阅[使用 SSH 登录](https://docs.qualcomm.com/bundle/publicresource/topics/80-70018-254/how_to.html#use-ssh)。
Note: 如果 SSH 已启用，则可以跳过此步骤。
- 从 Linux 主机推送模型文件。

        scp <model_filename> root@<IP address of target device>:/etc/modelsCopy to clipboard
- 请注意，[下载的脚本](https://docs.qualcomm.com/doc/80-70018-50SC/topic/download-model-and-label-files.html)会将示例 video.mp4 视频下载到 /etc/media 目录。如果您使用的是自定义视频，请确保将视频推送到 /etc/media，并在应用程序的 config.JSON 文件中更新文件路径。
- 使用 HDMI 端口将显示器连接到设备。有关说明，请参见[设置 HDMI 显示器](https://docs.qualcomm.com/bundle/publicresource/topics/80-70018-18/samples.html)。
- 启用显示器：

        export XDG_RUNTIME_DIR=/dev/socket/weston && export WAYLAND_DISPLAY=wayland-1Copy to clipboard

如果启用摄像头或显示器时遇到问题，请参阅[摄像头故障排除](https://docs.qualcomm.com/bundle/publicresource/topics/80-70018-17/troubleshooting.html)和[显示器故障排除](https://docs.qualcomm.com/bundle/publicresource/topics/80-70018-18/debug.html)。

## 运行应用程序

关于示例模型和标签文件，请参阅[示例模型和标签文件](https://docs.qualcomm.com/doc/80-70018-50SC/topic/gst-ai-parallel-inference.html#gst-ai-parallel-inference__section_pnn_hmb_4dc)。

进入 SSH shell 并将 YOLO-NAS 标签文件复制至 YOLOv8：

cp /etc/labels/yolonas.labels /etc/labels/yolov8.labelsCopy to clipboard

请使用以下语法，运行该应用程序：

    gst-ai-parallel-inference [OPTION?]Copy to clipboard

- 使用主摄像头和辅助摄像头的输入运行：

        gst-ai-parallel-inference -c 0Copy to clipboard

        gst-ai-parallel-inference -c 1Copy to clipboard
- 使用文件的输入运行：

        gst-ai-parallel-inference -s /etc/media/video.mp4Copy to clipboard
- 使用 RTSP 流的输入运行：

        gst-ai-parallel-inference --rtsp-ip-port="rtsp://<ip-address-of-device>/live.mkv"Copy to clipboard
- 使用视频流的输入以及 LiteRT 模型的常量运行所有四个推理：

        gst-ai-parallel-inference --file-path="/etc/media/video.mp4" --object-detection-constants="YOLOv8,q-offsets=<21.0, 0.0, 0.0>,q-scales=<3.0546178817749023, 0.003793874057009816, 1.0>;" --pose-detection-constants="Posenet,q-offsets=<8.0>,q-scales=<0.0040499246679246426>;" --segmentation-constants="deeplab,q-offsets=<0.0>,q-scales=<1.0>;" --classification-constants="Inceptionv3,q-offsets=<38.0>,q-scales=<0.17039915919303894>;"Copy to clipboard

要显示可用的帮助选项，请在 SSH shell 中运行以下命令：

    gst-ai-parallel-inference -hCopy to clipboard

要停止用例，请按 CTRL + C。

## 预期输出

执行四个并行推理后，结果将并排显示在屏幕上。

Figure : gst-ai-parallel-inference 应用程序的预期输出 
                
                ![](data:image/png;base64,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)

## Pipeline 流

该表列出了并行推理 pipeline 中使用的插件：| 插件 | 说明 |
| --- | --- |
| 摄像头源：[qtiqmmfsrc](https://docs.qualcomm.com/doc/80-70018-50SC/topic/qtiqmmfsrc.html) | <ul class="ul" id="gst-ai-parallel-inference__ul_zyl_gj1_mcc"><br>                                    <li class="li">从摄像头采集实时流。</li><br><br>                                    <li class="li">使用 tee 拆分流进行推理。</li><br><br>                                </ul> |
| 文件源：filesrc | <ul class="ul" id="gst-ai-parallel-inference__ul_z1z_x4f_w1c"><br>                                    <li class="li">使用 filesrc 采集视频流，然后使用 qtdemux 对视频流进行解复用。</li><br><br>                                    <li class="li">使用 tee 拆分流进行推理。</li><br><br>                                </ul> |
| RTSP 源：rtspsrc | <ul class="ul" id="gst-ai-parallel-inference__ul_vsj_2r4_tbc"><br>                                    <li class="li">使用 rtspsrc 采集 RTSP 流，然后使用 rtph264depay 进行视频提取。</li><br><br>                                    <li class="li">使用 tee 拆分流进行推理。</li><br><br>                                </ul> |
| h264parse | 渲染 H.264 视频。 |
| [v4l2h264dec](https://docs.qualcomm.com/doc/80-70018-50SC/topic/v4l2h264dec.html) | 解码视频 |
| [qtimlvconverter](https://docs.qualcomm.com/doc/80-70018-50SC/topic/qtimlvconverter.html) | <ol class="ol" id="gst-ai-parallel-inference__ol_kgt_hnq_nbc"><br>                                    <li class="li">在其接收端口上接收视频流。</li><br><br>                                    <li class="li">对流数据执行以下预处理。当模型需要浮点值作为输入时，会执行此预处理。<ol class="ol" type="a" id="gst-ai-parallel-inference__ol_drd_jnq_nbc"><br>                                            <li class="li">颜色转换</li><br><br>                                            <li class="li">缩放（放大或缩小）</li><br><br>                                            <li class="li">归一化</li><br><br>                                        </ol><br></li><br><br>                                </ol><br><br>                                <br>张量数据流用于 pipeline 后期的推理。 |
| [qtimltflite](https://docs.qualcomm.com/doc/80-70018-50SC/topic/qtimltflite.html) | <ol class="ol" id="gst-ai-parallel-inference__ol_l2x_zjq_nbc"><br>                                    <li class="li">推理 runtime 在其接收端口上接收到张量数据后，会运行推理。</li><br><br>                                    <li class="li">生成一个张量数据流，并在其发送端口上显示推理结果。</li><br><br>                                </ol> |
| 后处理插件 | 处理来自任何目标检测、分类、姿态检测和分割模型的推理结果。<ul class="ul" id="gst-ai-parallel-inference__ul_pgb_wpq_nbc"><br>                                    <li class="li"><a href="https://docs.qualcomm.com/doc/80-70018-50SC/topic/qtimlvdetection.html">qtimlvdetection</a><ol class="ol" id="gst-ai-parallel-inference__ol_ol3_dky_kbc"><br>                                            <li class="li">将阈值应用于所选结果数。</li><br><br>                                            <li class="li">加载 YOLO-NAS 模块。 </li><br><br>                                            <li class="li">生成仅包含边界框的视频帧，这些框可以叠加在对象上，并将它们发送到 qtivcomposer 的接收端口。</li><br><br>                                        </ol><br></li><br><br>                                    <li class="li"><a href="https://docs.qualcomm.com/doc/80-70018-50SC/topic/qtimlvclassification.html">qtimlvclassification</a><ol class="ol" id="gst-ai-parallel-inference__ol_nt3_jqq_nbc"><br>                                            <li class="li">将阈值应用于所选的结果数。 </li><br><br>                                            <li class="li">加载 MobileNet 模块。</li><br><br>                                            <li class="li">将结果生成为带有分类标签的视频帧，并将它们发送到 qtivcomposer 的接收端口。</li><br><br>                                        </ol><br></li><br><br>                                    <li class="li"><a href="https://docs.qualcomm.com/doc/80-70018-50SC/topic/qtimlvpose.html">qtimlvpose</a><ol class="ol" id="gst-ai-parallel-inference__ol_znf_mqq_nbc"><br>                                            <li class="li">将阈值应用于所选的结果数。</li><br><br>                                            <li class="li">为各种姿态估计模型加载相应的模块。对于本节中描述的用例，qtimlvpose 加载 PoseNet 模块。 </li><br><br>                                            <li class="li">将结果生成为带绘制姿势的视频帧，并将它们发送到 qtivcomposer 的接收端口。</li><br><br>                                        </ol><br></li><br><br>                                    <li class="li"><a href="https://docs.qualcomm.com/doc/80-70018-50SC/topic/qtimlvsegmentation.html">qtimlvsegmentation</a>：将接收端口上接收到的推理张量转换为视频格式，由多媒体插件进行后续处理。</li><br><br>                                </ul> |
| [qtivcomposer](https://docs.qualcomm.com/doc/80-70018-50SC/topic/qtivcomposer.html) | <ol class="ol" id="gst-ai-parallel-inference__ol_dmb_2vr_lbc"><br>                                    <li class="li">将接收端口获取的内容组成帧。</li><br><br>                                    <li class="li">将包含这些组合帧的 GStreamer 缓存推送到其发送端口。</li><br><br>                                </ol> |
| [Waylandsink](https://docs.qualcomm.com/doc/80-70018-50SC/topic/waylandsink.html) | <ol class="ol" id="gst-ai-parallel-inference__ol_kjr_fvr_lbc"><br>                                    <li class="li">Waylandsink 将其接收端口上接收的视频流提交给 Weston。</li><br><br>                                    <li class="li">Weston 在本地显示器上渲染视频流。</li><br><br>                                </ol> |

**Parent Topic:** [运行 AI/ML 示例应用程序](https://docs.qualcomm.com/doc/80-70018-50SC/topic/ai-ml-sample-applications.html)

**Related Resources**  

- [使用 Python 进行并行推理](https://docs.qualcomm.com/doc/80-70018-50SC/topic/parallel-inference-using-python.html)
- [图像分类](https://docs.qualcomm.com/doc/80-70018-50SC/topic/gst-ai-classification.html)
- [目标检测](https://docs.qualcomm.com/doc/80-70018-50SC/topic/gst-ai-object-detection.html)
- [姿态检测](https://docs.qualcomm.com/doc/80-70018-50SC/topic/gst-ai-pose-detection.html)
- [图像分割](https://docs.qualcomm.com/doc/80-70018-50SC/topic/gst-ai-segmentation.html)

Last Published: Nov 12, 2025

[Previous Topic
使用 Python 和容器进行图像分割](https://docs.qualcomm.com/bundle/publicresource/80-70018-50SC/topics/image-segmentation-using-python.md) [Next Topic
多输入/输出目标检测](https://docs.qualcomm.com/bundle/publicresource/80-70018-50SC/topics/gst-ai-multi-input-output-object-detection.md)