Setup: GPU Tesla T4 DeepStream 5.0.1 container CUDA 10.2 TensorRT 7.0.0 and 7.1.3 Driver 450.51.06 Question Hi, TLDR: Image resizing method in pipeline yields different result with opencv/PIL, making it difficult to align offline train/eval and online deployment. Is it possible to reveal some detail of resizing of DeepStream implementation? Details: For some time we had deployed a detector engine with deepstream pipeline. When dealing with the (mis)detection results, it is noticed that different image resize implementation could introduce noticeable classification error in our case. And we couldn’t get the detail of image resizing in nvvideoconvert or nvstreammux. nvvideoconvert has interpolation-method options, we tried nearest and bilinear as others are seldom used in production. nvstreammux has no such options, but it seems like nearest (which is dGPU default) Some experiment suggests the resizing operation produces different result with other common libs, e.g. opencv or PIL, which are used in offline train/eval. We do know the fact that opencv and PIL themselves don’t align with each other. But without further knowledge we could not choose or implement proper resizing library to unify the preprocessing in all environments. By digging into nvinfer we found the image buffer transformation are done by NvBufSurfTransform with default interpolation method (nearest). It also seems the interpolation-method directly maps to NvBufSurfTransformInter_XXX options. However the color conversion (NV12 to BGRA in our case) and resizing are done in the same function call to NvBufSurfTransform. Are they implemented separately or separable? libnvbufsurftransform.so depends on libnppig.so which provides resizing functions, but not libnppicc.so which provides NV12ToBGR conversions etc. Is the resizing done by NPP? Is there corresponding color conversion function in NPP? Specifically we are interested in aligning the following options: cv::INTER_NEAREST or PIL.Image.NEAREST, with NvBufSurfTransformInter_Nearest cv::INTER_LINEAR or PIL.Image.BILINEAR, with NvBufSurfTransformInter_Bilinear And if possible, may nvstreammux have interpolation-method option added in future, since it could perform image resize/resample when batching. Thanks.

After confirmed with the development team, the interpolation-method option with nvstreammux will be available with the next version deepstream sdk. We will announce the new version when it is available. Besides, if the input sources are in the same resolution, you can set nvstreammux “width” and “height” to the same as original video resolution, and the nvvideoconvert can be used after nvstreammux for scaling.

Current nvstreammux is using “NvBufSurfTransformInter_Bilinear” method for scaling. [image] vvivvy1999: And if possible, may nvstreammux have interpolation-method option added in future, since it could perform image resize/resample when batching. We will review your request and investigate it.

Hi @Fiona.Chen, thanks for the good message. Yes, in current pipeline we set nvvideoconvert before nvstreammux to rescale individual buffers which should work as configurable. However it seems a loss of performance if we further divide the NV12 to BGRA conversion and resizing into two separate nvvideoconverts then play with the resizing alone. After some experiment it seems with proper setup, nppiResizeSqrPixel series could almost align result with opencv. So if you could confirm that function found in libnvbufsurftransform.so is used as the resizing step, the problem will be mostly solved. Edit: nppiResizeSqrPixel details for resizing 1920x1080 grayscale buffer into 960x544 use xFactor = 960.0 / 1920.0, yFactor = 544.0 / 1080.0 xShift = yShift = 0.5, interpolation = NPPI_INTER_NN cv::INTER_NEAREST, identical result xShift = yShift = 0.0, interpolation = NPPI_INTER_LINEAR cv::inter_LINEAR, with some pixels value +/- 1

[image] vvivvy1999: Edit: nppiResizeSqrPixel details for resizing 1920x1080 grayscale buffer into 960x544 use xFactor = 960.0 / 1920.0, yFactor = 544.0 / 1080.0 xShift = yShift = 0.5, interpolation = NPPI_INTER_NN cv::INTER_NEAREST, identical result xShift = yShift = 0.0, interpolation = NPPI_INTER_LINEAR cv::inter_LINEAR, with some pixels value +/- 1 So you are talking about opencv CUDA version?

No, just cpu version. CUDA version is not tested.

We use cudaTexture for bilinear and nearest interpolations, the other interpolations use NPP.

Thanks very much. We’ll try about cudaTexture.

About the resize method in nvvideoconvert/nvstreammux

Accelerated Computing Intelligent Video Analytics DeepStream SDK

Fiona.Chen January 20, 2021, 7:19am 6

So you are talking about opencv CUDA version?

Difference between predictions of exported TensorRT engine and PyTorch pth models

Topic		Replies	Views
Nvstreammux resize method in DeepStream 8 DeepStream SDK deepstream	6	180	January 19, 2026
Resize interpolation in DeepStream 3 DeepStream SDK	3	1028	May 29, 2019
NvBufSurfTransform Interpolation Method Issue DeepStream SDK deepstream	5	151	February 3, 2026
Interpolation method in Gst-nvstreammux not working DeepStream SDK	10	1703	July 18, 2022
Accuracy decreased when using deepstream pipeline DeepStream SDK	7	1755	June 18, 2021
Deepstream NVInfer vs python resize preprocessing differences DeepStream SDK deepstream	13	485	September 1, 2025
Nvvideoconvert different resolutions support (input and output) DeepStream SDK	9	952	October 18, 2022
Scaling in jetson nano DeepStream SDK	6	1127	May 22, 2020
Query regarding deepstream interpolation methods DeepStream SDK deepstream61	2	553	December 7, 2023
Scaling in nvinfer DeepStream SDK	8	901	December 13, 2023

About the resize method in nvvideoconvert/nvstreammux

Related topics