Deepstream-app

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
**• DeepStream Version5.0
**• JetPack Version (valid for Jetson only)4.4
• 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)

How do you get the pixel value (RGB) for each frame in DeepStream-App, like

[[[255 252 253]
[255 252 253]
[255 252 253]
…
[ 90 90 90]
[ 95 89 86]
[ 96 90 87]]
[[255 252 253]
[255 252 253]
[255 252 253]
…
[ 91 91 91]
[ 96 90 87]
[ 96 90 87]]
[[255 252 253]
[254 251 252]
[255 252 253]
…
[ 95 89 93]
[ 93 91 91]
[ 94 92 93]]
…
[[ 41 45 34]
[ 46 50 39]
[ 47 49 39]
…
[ 60 57 63]
[ 63 57 66]
[ 66 59 68]]
[[ 50 47 37]
[ 52 49 38]
[ 48 47 38]
…
[ 63 61 59]
[ 61 61 63]
[ 64 63 66]]
[[ 54 51 41]
[ 50 47 37]
[ 46 46 37]
…
[ 71 69 67]
[ 64 63 66]
[ 64 63 66]]]

3

What do you mean? Do you want to print the channel value for every pixel with deepstream-app? If so, we don’t support such function in deepstream-app.

But the deepstream-app sample codes are open source. You can modify and add you own processing as you like.

4

Thank you very much for your reply. Is there any pixel information about host_rgb_buf in the structure _GstDsExample in the gstdsexample.h, and if so, how to call it?

5

In deepstream, all frame buffers(which contain all pixel data and frame information) are encapsulated in NvBufSurface. NVIDIA DeepStream SDK API Reference: Main Page
The NvBufSurfaceParams inside NvBufSurface (NVIDIA DeepStream SDK API Reference: Main Page) can show all color format and buffer address information .

You can find NvBufSurface has already been got in gst_dsexample_transform_ip in gstdsexample.cpp.

The definition of NvBufSurface and related structures can be found in /opt/nvidia/deepstream/deepstream-5.0/sources/includes/nvbufsurface.h

6

The host_rgb_buf format is a 3 channels, 8 bit per channel RGB buffer. You can know the format with the codes of gst_dsexample_start(). But it is actually not in use. It is just a sample of how to allocate CUDA buffer. The actual data in the buffer is not decided yet.