Subsampling in DeepStream

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
T4
• DeepStream Version
5.0
• JetPack Version (valid for Jetson only)
NA
• TensorRT Version
7.0.0.11
• NVIDIA GPU Driver Version (valid for GPU only)
440.64.00
• Issue Type( questions, new requirements, bugs)

Is there an option for subsampling input images in any nvidia DeepStream plugin? Say in nvvideoconvert or nvstreammux or nvinfer? If not, is there a plan to add such an option.

By subsampling what I mean is – if there are 30 images coming in per second and the NN batch size is 10, we can just set the subsample factor to 3 so that we can run one inference per second.

Are you talking about the case with one stream which is 30 FPS?

1 stream with 30 fps subsampled at 1 image processed per 3 images input. And this could extend to many the many-streams case too.

If there is only one stream, deepstream can handle one batch per 10 images.

For multiple streams, deepstream can not work in this way. Frequently Asked Questions — DeepStream 5.1 Release documentation

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.