How to integrate deepstream_parallel_inference_app in deepstream_app.c

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

• Hardware Platform ( GPU: Nvidia GeForce RTX 3060)
• DeepStream Version: 6.3 docker container
• NVIDIA GPU Driver Version: 535.129.03
• Issue Type( questions)

I tried this repo https://github.com/NVIDIA-AI-IOT/deepstream_parallel_inference_app for a parallel inference app using 2 models. I want to integrate the bodypose_yolo with deepstream_app.c in deepstream app. Is that possible to do so? If yes, could you please help me out with or even share a sample code I could refer to?

3

What do you mean by this request?

The GitHub - NVIDIA-AI-IOT/deepstream_parallel_inference_app: A project demonstrating how to use nvmetamux to run multiple models in parallel. is already developed with the same style as deepstream-app. You can configure different pipelines by the configuration YAML file.

4

I want to integrate the parallel app branches in deepstream_app.c. Like the bodypose2D and yolov4 detection inference going on parallel to each other in the same pipeline in deepstream_app.c instead of this parallel-app repo

5

The sample GitHub - NVIDIA-AI-IOT/deepstream_parallel_inference_app: A project demonstrating how to use nvmetamux to run multiple models in parallel. is to show you how to run models in parallel. The sample is in the same style of deepstream-app. And they are all open source, you can compare and understand the differences between the two samples.

6

Which deepstream test app is it similar to? I couldn’t see similar function between deepstream_app.c and deepstream_parallel_infer_app.cpp. For example, where is the write_kitti_output written here to save the txt files?

7

They are different samples while they are written by the same interfaces in /opt/nvidia/deepstream/deepstream/sources/apps/apps-common. If you want any function in deepstream-app, you can port the same function to GitHub - NVIDIA-AI-IOT/deepstream_parallel_inference_app.

8

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks

9

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