# mt-ds-app **Repository Path**: taj5/mt-ds-app ## Basic Information - **Project Name**: mt-ds-app - **Description**: 运行在nvidia dGPU和T4L平台,多路取流、解码、推理、后处理、编码、推流应用。 - **Primary Language**: Unknown - **License**: MulanPSL-2.0 - **Default Branch**: master - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 0 - **Forks**: 0 - **Created**: 2023-07-03 - **Last Updated**: 2025-09-09 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README # mt-ds-app ## 介绍 运行在nvidia dGPU和T4L平台,多路取流、解码、推理、后处理、编码、推流应用。 ## 软件架构 ```bash mt-ds-app ├── apps-common # 公用代码 │ ├── includes │ └── src ├── configs # 配置 ├── classification # 分类pipeline ├── detection # 检测pipeline ├── segmentation # 分割pipeline ├── Makefile ├── models # 模型 ├── README.md └── streams # 测试视频 ``` ## 安装教程 [deepstream-6.2 dGPU Setup](https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_Quickstart.html#dgpu-setup-for-ubuntu) 其他依赖组件 - GStreamer-1.0 - GStreamer-1.0 Base Plugins - GStreamer-1.0 gstrtspserver - X11 client-side library - Glib json library - json-glib-1.0 - yaml-cpp ``` bash # 安装 sudo apt-get install libgstreamer-plugins-base1.0-dev libgstreamer1.0-dev \ libgstrtspserver-1.0-dev libx11-dev libjson-glib-dev libyaml-cpp-dev ``` ## 使用说明 ```bash # 运行分类程序 $ ./cls # 运行检测程序 $ ./det detection/config/config.yml # docker镜像启动容器 $ docker run --gpus all -it -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=$DISPLAY --net=host bash ``` 当视频文件为HEVC/265编码时,source插件可能不支持,需要转为AVC/264编码格式,使用如下命令 ```bash $ ffmpeg -i hevc.mp4 -c:v libx264 avc.mp4 ``` ## 特技 ### cuda延迟加载 cuda lazy loading can reduce device memory usage and speed up TensorRT initializion.[Ref](https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#lazy-loading) ```bash # set this to enable lazing loading export CUDA_MODULE_LOADING=LAZY ``` ### 安装rtspclientsink rtspclientsink plugin is not part of the basic three plugin packages. You need to install it separately. To install it, run the following command: ``` sudo apt install gstreamer1.0-rtsp ``` ## 开发计划 - 构建分类模型的IVA应用 classification - 构建检测模型的IVA应用 ds-detection - 构建分割模型的IVA应用 ds-segmentation - 检测+跟踪+分类IVA应用 ds-app