How to rotate the video before sending it to inference?

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

For a realtime detection application, I want to rotate the video input first, and then send it to inference. Is it possible to implement it by Deepstream?

4

Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
• The pipeline being used

5

DeepStream supports this operation, you can refer to deepstream_custom_preprocessing_app to configure the transformation/rotation matrix.

However, could you explain why this is necessary? Perhaps there are other ways to do it?

This repository has only been tested on DS-8.0. If you are not using DS-8.0, please try porting it yourself.

6

Thank you very much for the help.

This seems a CPU based solution and depends on DS 8.0, and we are still using DS 7.1. We will currently using OpenCV to rotate the video, and will try this solution later.

7

This is a CUDA solution that combines linear transformation and normalization; it’s not a CPU solution. This code should be easily portable to DS-7.1.

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.