# 视频超分辨率

Source: [https://docs.qualcomm.com/doc/80-70015-50SC/topic/video-super-resolution.html](https://docs.qualcomm.com/doc/80-70015-50SC/topic/video-super-resolution.html)

**gst-ai-superresolution** 程序可以使用低分辨率输入生成高分辨率视频帧。

下图显示了一个接受来自文件源的视频流输入，使用 TFLite runtime 超分辨率模块处理视频流，并显示输出的 pipeline。

有关 pipeline 中使用的插件的信息，可参见[Pipeline 流程](https://docs.qualcomm.com/doc/80-70015-50SC/topic/video-super-resolution.html#video-super-resolution__section_kkk_xhz_lcc)。

Figure : gst-ai-superresolution pipeline
            ![](data:image/jpeg;base64,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)

## 前提条件

- 要运行应用程序，需将模型和标签文件推送到设备。有关下载模型的信息，可参见以下内容：
    - [下载 Qualcomm Neural Processing SDK 的模型和标签文件](https://docs.qualcomm.com/doc/80-70015-50SC/topic/ai-ml-sample-applications.html#ai-ml-sample-applications__section_chl_dgz_scc)
    - [从 AI Hub 下载 TFLite 的模型和标签文件](https://docs.qualcomm.com/doc/80-70015-50SC/topic/ai-ml-sample-applications.html#ai-ml-sample-applications__section_fsl_lgz_scc)

    该应用程序支持 Qualcomm Neural Processing SDK 和 TFLite 模型。
- 要访问主机设备，应启用 SSH。相关说明，可参见 [SSH 使用指南](https://docs.qualcomm.com/bundle/publicresource/topics/80-70015-254/how_to.html#how-to-ssh-)
- 进入 SSH shell 并运行用例：

        ssh root@<ip-addr of the target device>Copy to clipboard
- 启用显示器：

        export XDG_RUNTIME_DIR=/dev/socket/weston && export WAYLAND_DISPLAY=wayland-1Copy to clipboard
- 从主机推送文件：

        scp <filename> root@<IP address of target device>:/opt/Copy to clipboard

- 将 video.mp4 文件推送到 opt 文件夹中。

## 用例

- 使用 Wayland 显示器显示输出：

        gst-ai-superresolution --input-file=/opt/video.mp4 --displayCopy to clipboard
- 使用 filesink 将输出保存到文件：

        gst-ai-superresolution --input-file=/opt/video.mp4 --output-file=/opt/out.mp4Copy to clipboard
- 使用 TFLite runtime 运行包含常量的模型：

        gst-ai-superresolution --input-file=/opt/video1.mp4 --model=/opt/quicksrnetsmall_quantized.tflite --constants="srnet,q-offsets=<0.0>,q-scales=<1.0>;"Copy to clipboard

## 预期输出

输出显示在 HDMI 监视器上。

Figure : VSR 的预期输出
                ![](data:image/png;base64,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)

## Pipeline 流程

下表列出了视频超分辨率 pipeline 中使用的插件：| 插件 | 说明 |
| --- | --- |
| filesrc | 采集视频流，并使用 tee 拆分流以进行推理。 |
| [qtimlvconverter](https://docs.qualcomm.com/doc/80-70015-50SC/topic/qtimlvconverter.html) | AI 处理流用于预处理：<ol class="ol" id="video-super-resolution__ol_j34_ddg_q1c"><br>                                    <li class="li">在其接收设备接插口上接收视频流。</li><br><br>                                    <li class="li">对流数据执行以下预处理。当模型使用浮点值输入时，会执行此预处理。<ol class="ol" type="a" id="video-super-resolution__ol_m5z_cpr_lbc"><br>                                            <li class="li">色彩转换</li><br><br>                                            <li class="li">缩放（向上或向下）</li><br><br>                                            <li class="li">归一化</li><br><br>                                        </ol><br></li><br><br>                                    <li class="li">将预处理的视频流转换为其源接插口上的张量数据。</li><br><br>                                </ol><br><br>张量数据用于 pipeline 后期的推理。 |
| [qtimltflite](https://docs.qualcomm.com/doc/80-70015-50SC/topic/qtimltflite.html) | 在 TFLite runtime 上运行，并使用 quicksrnetsmall\_quantized 模型。<ol class="ol" id="video-super-resolution__ol_lwl_521_mcc"><br>                                    <li class="li">推理 runtime 在其接收设备接插口上接收张量数据。</li><br><br>                                    <li class="li">runtime 运行推理。</li><br><br>                                    <li class="li">生成一个张量数据，并在其源接插口上显示推理结果。</li><br><br>                                </ol><br>用于处理推理的后处理插件来自 Midasv2 模型。 |
| [qtimlvsuperresolution](https://docs.qualcomm.com/doc/80-70015-50SC/topic/qtimlvsuperresolution.html) | 处理来自任何超分辨率模式的推理结果。<ol class="ol" id="video-super-resolution__ul_w2c_xf1_mcc"><br>                                    <li class="li">加载 SRNet 模块。</li><br><br>                                    <li class="li">将结果生成为视频帧。</li><br><br>                                    <li class="li">将它们发送到 qtivcomposer 的接收设备接插口。</li><br><br>                                </ol><br><br>为 INT8 模型添加以下常量：<br><ul class="ul" id="video-super-resolution__ul_eqr_zf1_mcc"><br>                                    <li class="li">q-offset</li><br><br>                                    <li class="li">q-scale</li><br><br>                                </ul> |
| [qtivcomposer](https://docs.qualcomm.com/doc/80-70015-50SC/topic/qtivcomposer.html) | <ol class="ol" id="video-super-resolution__ol_dmb_2vr_lbc"><br>                                    <li class="li">从其接收设备接插口获取内容然后组成帧。</li><br><br>                                    <li class="li">将包含这些组合帧的 GStreamer 缓冲区推送到其源接插口。</li><br><br>                                </ol> |
| [Waylandsink](https://docs.qualcomm.com/doc/80-70015-50SC/topic/waylandsink.html) | <ol class="ol" id="video-super-resolution__ol_kjr_fvr_lbc"><br>                                    <li class="li">Waylandsink 将在其接收设备接插口上接收的视频流提交给 Weston。</li><br><br>                                    <li class="li">Weston 在本地显示器上呈现视频流。</li><br><br>                                </ol> |

**Parent Topic:** [AI/ML 示例程序](https://docs.qualcomm.com/doc/80-70015-50SC/topic/ai-ml-sample-applications.html)

Last Published: Nov 11, 2025

[Previous Topic
视频单目深度](https://docs.qualcomm.com/bundle/publicresource/80-70015-50SC/topics/mono-depth-from-video.md) [Next Topic
多流推理](https://docs.qualcomm.com/bundle/publicresource/80-70015-50SC/topics/multistream-inference.md)