Deepstream-pose-estimation-app does not proceed

• Hardware Platform (Jetson / GPU): Jetson TX2
• DeepStream Version: 5.0.1
• JetPack Version (valid for Jetson only): 4.4
• TensorRT Version: 7.1.3.0
• CUDA Version: 10.2

I am testing DeepStream Human Pose Estimation.
I followed the steps below

:/opt/nvidia/deepstream/deepstream-5.0/sources/apps/sample_apps/deepstream_pose_estimation$ sudo ./deepstream-pose-estimation-app movie/dance.mp4 output
[sudo] password for nvidia: 
Now playing: movie/dance.mp4
Opening in BLOCKING MODE 
Opening in BLOCKING MODE 
0:00:06.783499989  6930   0x5587276580 INFO                 nvinfer gstnvinfer.cpp:619:gst_nvinfer_logger:<primary-nvinference-engine> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:1701> [UID = 1]: deserialized trt engine from :/opt/nvidia/deepstream/deepstream-5.0/sources/apps/sample_apps/deepstream_pose_estimation/pose_estimation.onnx_b1_gpu0_fp16.engine
INFO: [Implicit Engine Info]: layers num: 3
0   INPUT  kFLOAT input.1         3x224x224       
1   OUTPUT kFLOAT 262             18x56x56        
2   OUTPUT kFLOAT 264             42x56x56        

0:00:06.783667797  6930   0x5587276580 INFO                 nvinfer gstnvinfer.cpp:619:gst_nvinfer_logger:<primary-nvinference-engine> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:1805> [UID = 1]: Use deserialized engine model: /opt/nvidia/deepstream/deepstream-5.0/sources/apps/sample_apps/deepstream_pose_estimation/pose_estimation.onnx_b1_gpu0_fp16.engine
0:00:06.822586209  6930   0x5587276580 INFO                 nvinfer gstnvinfer_impl.cpp:313:notifyLoadModelStatus:<primary-nvinference-engine> [UID 1]: Load new model:deepstream_pose_estimation_config.txt sucessfully
Running...
NvMMLiteOpen : Block : BlockType = 261 
NVMEDIA: Reading vendor.tegra.display-size : status: 6 
NvMMLiteBlockCreate : Block : BlockType = 261 

However, It’s been 3 hours since the bottom line and it’s still not working.
What was the cause?

Thank you.

Hi,

Do you connect a display to the Jetson.
You should find the output on the display already.

Thanks.

Yes. I connect it.
But, I can’t find the output.

And the output directory is empty.

Similarly, for deepstream-test1, there was no display output.

On the other hand, when I ran deepstream-app -c source30_1080p_dec_infer-resnet_tiled_display_int8.txt, the video was output to the display.

nvidia@nvidia-desktop:/opt/nvidia/deepstream/deepstream-5.0/samples/configs/deepstream-app$ deepstream-app -c source30_1080p_dec_infer-resnet_tiled_display_int8.txt

Hi,

Sorry that deepstream-pose-estimation-app should be a video output rather than display.
Do you find Pose_Estimation.mp4 file in the output folder?

Thanks.

Thank you for your reply.

I found it, but the size of it is 0.

nvidia@nvidia-desktop:/opt/nvidia/deepstream/deepstream-5.0/sources/apps/sample_apps/deepstream_pose_estimation$ ls -la output
total 8
drwxrwxrwx 2 nvidia nvidia 4096 May 11 18:15 .
drwxrwxrwx 7 root   root   4096 May 11 15:47 ..
-rw-r--r-- 1 root   root      0 May 11 18:15 Pose_Estimation.mp4

Sorry, I changed the video format to h264 and it worked fine.