Get TensorMeta given GIE_ID when there are multi PGIE/SGIE

Accelerated Computing Intelligent Video Analytics DeepStream SDK
,
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) GPU
• DeepStream Version 6.1
• JetPack Version (valid for Jetson only)
• TensorRT Version 8.5.2
• NVIDIA GPU Driver Version (valid for GPU only) 535
• Issue Type( questions, new requirements, bugs) questions

I want to run deepstream with multi PGIE/SGIE and then get their tensor_meta to run post-processing. However, when I look at the example deepstream_ssd_parser, it seems there is no such option. Do you have any suggestions?
For example, I want to run 2 PGIE models person detection ID1 and car detection ID2, then run embedding extraction SGIE only on person bboxes?

3

There is no update from you for a period, assuming this is not an issue any more. Hence we are closing this topic. If need further support, please open a new one. Thanks.
most cases are pgie+sgie. there is no mutiple pgie sample. you can set process-mode to 1 in the second pgie’s configuration file if need.
there is a pgie+sgie sample. the pgie detects the person, and the sgie gets the embedding vector of persions.

4

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