nvidia示例

机器学习 960 字 次

nvidia示例。

安装 deepstream SDK

方法一：使用 sdkmanager

在sdkmanager中选择deepstream，按刷机的方式安装即可。

方法二：单独下载

下载： (要登录）

安装：

1
2
3
4
- CUDA (10.2)
- TensorRT (7.1+)
- OpenCV (3.3.1)
- VisionWorks (1.6)

安装工具：

1
2
3
4
5
6
7
8
9
10
11
$ sudo apt-get install \
    libssl1.0.0 \
    libgstreamer1.0-0 \
    gstreamer1.0-tools \
    gstreamer1.0-plugins-good \
    gstreamer1.0-plugins-bad \
    gstreamer1.0-plugins-ugly \
    gstreamer1.0-libav \
    gstreamer1.0-alsa \
    libgstrtspserver-1.0-0 \
    libjansson4

安装：

1
./install.sh
1
2
3
4
5
md5sum deepstream_sdk_v5.0.1_jetson.tbz2 
5c60a89924535129ec3b4997ebbc0f83  deepstream_sdk_v5.0.1_jetson.tbz2

sudo nvpmodel -m 0 
sudo jetson_clocks

安装环境后，检查：

1
2
3
4
5
6
7
8
# deepstream-app --version-all 
deepstream-app version 5.0.0
DeepStreamSDK 5.0.0
CUDA Driver Version: 10.2
CUDA Runtime Version: 10.2
TensorRT Version: 7.1
cuDNN Version: 8.0
Dewarper: not found
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
模型文件：/opt/nvidia/deepstream/deepstream-5.0/samples/models

示例文档：https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_ref_app_github.html  
示例地址：https://github.com/NVIDIA-AI-IOT/deepstream_reference_apps。克隆其到 /opt/nvidia/deepstream/deepstream_reference_apps。  

安装依赖：
查看：/opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-app/README  
命令：   
sudo apt-get install libgstreamer-plugins-base1.0-dev libgstreamer1.0-dev \
   libgstrtspserver-1.0-dev libx11-dev libjson-glib-dev
编译：make （验证SDK、依赖是否完备）。  

将 anomaly 拷贝到 /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/ ：
cp  /opt/nvidia/deepstream/deepstream_reference_apps/anomaly/ /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/ 

进入目录：
cd /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/anomaly
编译库：
cd plugins/gst-dsdirection/
make && sudo make install

编辑 ./anomaly/apps/deepstream-anomaly-detection-test/dsanomaly_pgie_config.txt 文件。

测试资源：
/opt/nvidia/deepstream/deepstream/samples/streams
/opt/nvidia/deepstream/deepstream/samples/streams/sample_720p.h264


gst-launch-1.0 filesrc location = samples/streams/sample_1080p_h264.mp4 ! qtdemux ! h264parse ! nvv4l2decoder ! m.sink_0 \
nvstreammux name=m batch-size=1 width=1280 height=720 ! nvinfer config-file-path= samples/configs/deepstream-app/config_infer_primary.txt  \
! nvof ! tee name=t ! queue ! nvofvisual ! nvmultistreamtiler width=1920 height=1080 !  nvegltransform ! nveglglessink sync=0 t. ! queue ! dsdirection ! \
nvmultistreamtiler width=1920 height=1080 ! nvvideoconvert ! nvdsosd ! nvegltransform ! nveglglessink sync=0

gst-launch-1.0 -e nvstreammux name=mux batch-size=2 width=1920 height=1080 ! nvinfer config-file-path=/opt/nvidia/deepstream/deepstream-5.0/samples/configs/deepstream-app/config_infer_primary.txt batch-size=2  \
! nvstreamdemux name=demux filesrc location=/opt/nvidia/deepstream/deepstream-5.0/samples/streams/sample_1080p_h264.mp4 ! qtdemux ! h264parse ! nvv4l2decoder ! queue \
! mux.sink_0 filesrc location=/opt/nvidia/deepstream/deepstream-5.0/samples/streams/sample_720p.h264 ! h264parse ! nvv4l2decoder ! queue ! mux.sink_1 demux.src_0 ! "video/x-raw(memory:NVMM), format=NV12" \
! queue ! nvvideoconvert ! "video/x-raw(memory:NVMM), format=RGBA" ! nvdsosd ! nvvideoconvert ! nveglglessink demux.src_1 ! queue ! "video/x-raw(memory:NVMM), format=NV12" ! queue ! nvvideoconvert ! "video/x-raw(memory:NVMM), format=RGBA" ! nvegltransform ! nveglglessink

出错（在图形界面中运行也一样）：
nvbuf_utils: Could not get EGL display connection
WARNING: erroneous pipeline: no element "nvstreammux"

2个示例：
cd /opt/nvidia/deepstream/deepstream_reference_apps/anomaly/apps/deepstream-anomaly-detection-test
./deepstream-anomaly-detection-app  /opt/nvidia/deepstream/deepstream/samples/streams/sample_720p.h264

cd /opt/nvidia/deepstream/deepstream_reference_apps/back-to-back-detectors
./back-to-back-detectors /opt/nvidia/deepstream/deepstream/samples/streams/sample_720p.h264

出错及解决：

1
2
3
4
5
6
提示：
nvbuf_utils: Could not get EGL display connection：
临时设置环境环境： export DISPLAY=:0

One element could not be created. Exiting.

查看插件：

1
2
3
4
5
6
7
8
9
10

filesrc

gst-inspect-1.0 -a |grep -i NVIDIA

$ gst-inspect-1.0  nvarguscamerasrc
有信息输出

$ gst-inspect-1.0  nvstreammux
No such element or plugin 'nvstreammux'
1
2
3
sudo docker run -it --rm --net=host --runtime nvidia  -e DISPLAY=$DISPLAY -w /opt/nvidia/deepstream/deepstream-5.0 -v /tmp/.X11-unix/:/tmp/.X11-unix -v /opt/nvidia/deepstream/deepstream_reference_apps:/opt/nvidia/deepstream/deepstream_reference_app nvcr.io/nvidia/deepstream-l4t:5.0.1-20.09-samples

 sudo docker run -it --rm --net=host --runtime nvidia  -e DISPLAY=$DISPLAY -v /tmp/.X11-unix/:/tmp/.X11-unix  nvcr.io/nvidia/deepstream-l4t:5.0.1-20.09-base  bash

安装指南：

  • 本文作者：李迟
  • 版权声明：原创文章，版权归署名作者，转载建议注明出处（当然不注明亦可）。
  • 本文链接：/ml/nvidia-sample.html