Inaccurate Input Preprocessing Before Feeding Into Model

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) GPU
• DeepStream Version 6.3
• Issue Type( questions, new requirements, bugs) questions

Hi, I am trying to implement a custom classification model and debug the output differences between the DeepStream pipeline and the ONNX model. When I save the pre-processed image before inference, it appears to be split into an NxN kernel (not smooth) (image 1) compared to the original image (image 2). As a result, the output probabilities differ from the ONNX pipeline. Could you tell me why this happens and how it affects the pipeline?

image
image526×627 20.8 KB

image
image558×653 22.5 KB

3

You can refer to our FAQ first to tune some parameters to check the effect.

4

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

5

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