Action Recognition Net Integration with Deepstream

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform GPU
• DeepStream Version 7.0
• TensorRT Version 8.6.1.6
• NVIDIA GPU Driver Version 535.247.01
**• How to reproduce the issue ? **
The sample deepstream 3d action recognition app which is throwing an error while running the pipeline on sample videos. The error is:
/opt/nvidia/deepstream/deepstream-7.0/sources/apps/sample_apps/deepstream-3d-action-recognition$ ./deepstream-3d-action-recognition -c deepstream_action_recognition_config.txt
num-sources = 4
Now playing: file:///opt/nvidia/deepstream/deepstream/samples/streams/sample_walk.mov, file:///opt/nvidia/deepstream/deepstream/samples/streams/sample_ride_bike.mov, file:///opt/nvidia/deepstream/deepstream/samples/streams/sample_run.mov, file:///opt/nvidia/deepstream/deepstream/samples/streams/sample_push.mov,
WARNING: ../nvdsinfer/nvdsinfer_model_builder.cpp:1494 Deserialize engine failed because file path: /opt/nvidia/deepstream/deepstream-7.0/sources/apps/sample_apps/deepstream-3d-action-recognition/./resnet18_3d_rgb_hmdb5_32.etlt_b4_gpu0_fp16.engine open error
0:00:08.512995051 111901 0x5bfa5e6c60f0 WARN nvinfer gstnvinfer.cpp:679:gst_nvinfer_logger: NvDsInferContext[UID 1]: Warning from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:2083> [UID = 1]: deserialize engine from file :/opt/nvidia/deepstream/deepstream-7.0/sources/apps/sample_apps/deepstream-3d-action-recognition/./resnet18_3d_rgb_hmdb5_32.etlt_b4_gpu0_fp16.engine failed
0:00:08.753454414 111901 0x5bfa5e6c60f0 WARN nvinfer gstnvinfer.cpp:679:gst_nvinfer_logger: NvDsInferContext[UID 1]: Warning from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:2188> [UID = 1]: deserialize backend context from engine from file :/opt/nvidia/deepstream/deepstream-7.0/sources/apps/sample_apps/deepstream-3d-action-recognition/./resnet18_3d_rgb_hmdb5_32.etlt_b4_gpu0_fp16.engine failed, try rebuild
0:00:08.753488670 111901 0x5bfa5e6c60f0 INFO nvinfer gstnvinfer.cpp:682:gst_nvinfer_logger: NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::buildModel() <nvdsinfer_context_impl.cpp:2109> [UID = 1]: Trying to create engine from model files
NvDsInferCudaEngineGetFromTltModel: Failed to open TLT encoded model file /opt/nvidia/deepstream/deepstream-7.0/sources/apps/sample_apps/deepstream-3d-action-recognition/./resnet18_3d_rgb_hmdb5_32.etlt
ERROR: ../nvdsinfer/nvdsinfer_model_builder.cpp:733 Failed to create network using custom network creation function
ERROR: ../nvdsinfer/nvdsinfer_model_builder.cpp:799 Failed to get cuda engine from custom library API
0:00:16.271412169 111901 0x5bfa5e6c60f0 ERROR nvinfer gstnvinfer.cpp:676:gst_nvinfer_logger: NvDsInferContext[UID 1]: Error in NvDsInferContextImpl::buildModel() <nvdsinfer_context_impl.cpp:2129> [UID = 1]: build engine file failed
free(): double free detected in tcache 2
Aborted (core dumped)

How to resolve this?

3

Have you modified the configuration file? 3d_action does not require a custom engine to generate functions

I can run it correctly on DS-7.0 and DS-7.1

4

I did make the changes as mentioned in the README file.
config_infer_primary_3d_action.txt (2.6 KB)
deepstream_action_recognition_config.txt (2.0 KB)
config_preprocess_3d_custom.txt (2.5 KB)
Find the Configuration files for the deepstream-app and action recognition net above.

5

This may be caused by incorrect installation. Try using the Docker image of DS7.1. The new version of DeepStream no longer supports etlt, but only ONNX.

export DISPLAY=:0 

xhost +

docker run --gpus all -it --rm --network=host --privileged -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=$DISPLAY -w /opt/nvidia/deepstream/deepstream-7.1 nvcr.io/nvidia/deepstream:7.1-gc-triton-devel
6

Thanks for your support the application is working when etlt file is replaced with the onnx file.

7

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.