Query regarding deepstream interpolation methods

[vino_mm]

• Hardware Platform (Jetson / GPU) Jetson
• DeepStream Version 6.1
• JetPack Version (valid for Jetson only) 5.0.2
• TensorRT Version 8.4.1
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs) Question

Please give the answer for both Jetson and GPU devices.

1. What is the default interpolation method used in nvstreammux?
2. What is the default interpolation method used in nvinfer scaling-filter?
3. Does NvBufSurfTransformInter_Nearest/NvBufSurfTransformInter_Bilinear runs on CPU or GPU?
4. Is the below the order and available interpolation methods?

[yuweiw]

1.Nearest
2.You can refer to our open source code: gstnvinfer.cpp

  nvinfer->transform_config_params.compute_mode = NvBufSurfTransformCompute_Default;
  nvinfer->transform_params.transform_filter = NvBufSurfTransformInter_Default;

3.They both run on GPU on dgpu platforms by default, but on jetson they both run on VIC by default.

4.Yes. And you can refer to our open source too. nvbufsurftransform.h

[vino_mm]

@yuweiw Thank you.