How to inference on multiple images with deepstream_python_apps

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Hello. I am trying to make the inference code on multiple images with deepstream_python_apps.

I am using the trafficCamNet on the Jetson AGX device.

Do anyone know how to do inference on multiple images with deepstream_python_apps?

Please give me any help.

Best Regards

3

does “multiple images” mean multiple jpg images?

you can use multifilesrc - multifilesrc

4

Thank you for your reply.

Yes, multipe .jpg or .png images.

By the way, your suggested link doesn’t contain the information for inference on multiple images using deepstream_python_apps.

Please let me know if I am wrong.

5

deepstream_python_apps includes multiple samples, which sample do you want?
If you are not sure, please look over the sample, and doc - https://docs.nvidia.com/metropolis/deepstream/dev-guide/index.html#page/DeepStream_Development_Guide/deepstream_Python_sample_apps.html#wwpID0EGHA

6

Thank you so much for your reply.

Since deepstream_python_apps only works on H264 video or RTSP, I created the H264 video with multiple png images.

It was my solution. :(

7

you can replace the filesrc for H264 video with multifilesrc for multiple jpg files

8

Oh I will try and will reply with the results

Thank you so much.