Getting Started with GstCUDA Wrapper

Follow Us On

Share This Page

Facebook Twitter LinkedIn WhatsApp Email Copy Link

Close

GstCUDA
Features and Limitations
Supported Platforms
Evaluating GstCUDA
Building and Installation Guide
Project Structure
GstCUDA Framework
libGstCUDA API GstCUDABaseFilter GstCUDABaseMISO GstCUDABaseSIMO GstCUDABaseMIMO
GstCUDA Quick Prototyping Elements
cudafilter cudamux cudademux cudamimo
GstCUDA Algorithm Developer’s Guide
GstCUDA Add-Ons
cudadebayer
Examples
cudafilter cudafilter: NVMM direct mapping cudafilter: Unified memory allocator cudamux cudamux: NVMM direct mapping cudamux: Unified memory allocator cudadebayer opencv opencvwarp opencvfilter
GstCuda Wrapper (new!)
Introduction Getting Started Usability
Performance Profiling
Contact Us

Getting the code

GstCUDA wrapper source code is delivered after the purchase of a GstCUDA or a product that is GstCUDA-based.

For more information regarding GstCUDA and GstCUDA wrapper, don't hesitate to contact us:

Contact Us!

Installing dependencies

Meson building system

At the building system level:

• ninja-build
• python3-pip
• pkg-config
• meson

Please, install them using (in Debian-based systems):

sudo apt install -y \
     python3        \
     python3-pip    \
     ninja-build    \
     pkg-config

The process should be similar in other systems, not based on Debian.

GStreamer Installation

It is required to provide a custom GStreamer build since the default one included in the JetPack may not expose the CUDA library. The only necessary option is nvcodec for the gst-plugins-bad subproject.

meson setup builddir -Dgst-plugins-bad:nvcodec=enabled --prefix=/usr
ninja -C build
sudo ninja -C build install

Building the code

Once fulfilled the dependencies, you can run a default compilation with:

meson build --optimization 3 --prefix /usr -Denable-docs=disabled -Ddeveloper-mode=false
ninja -C build
sudo ninja -C build install

For additional customization, you may refer to the following table of options:

Project Option	Default Value	Possible Values	Description
developer-mode	true	[True, False]	Enable developer mode
enable-docs	enabled	[enabled, disabled, auto]	Build documentation with Doxygen
enable-docs-only	disabled	[enabled, disabled, auto]	Build documentation with Doxygen without compiling the project
gstcuda-backend	ridgerun	[ridgerun, gstreamer]	Select which GstCUDA implementation to use

Some additional options are required:

Configure Option	Description
--prefix /usr	Set the installation path of the library.
--optimization 3	Set the optimization level to the maximum

Testing the code

After installing GstCUDA Wrapper, it installs a new element capable of demonstrating the functionality.

Check element installation

First, check that the element is detected by GStreamer:

gst-inspect-1.0 cudawrapperfilter

It should show the inspect of the element.

Run the example element

For Jetpack < 7:

gst-launch-1.0 videotestsrc is-live=1 ! nvvidconv ! "video/x-raw(memory:NVMM)" ! rrcudawrapperfilter in-place=false ! fakesink

gst-launch-1.0 videotestsrc is-live=1 ! nvvidconv ! "video/x-raw(memory:NVMM)" ! rrcudawrapperfilter in-place=true ! fakesink

For Jetpack >= 7:

gst-launch-1.0 videotestsrc is-live=1 ! nvvidconv ! "video/x-raw(memory:NVMM)" ! cudaupload ! rrcudawrapperfilter in-place=false ! cudadownload ! fakesink

gst-launch-1.0 videotestsrc is-live=1 ! nvvidconv ! "video/x-raw(memory:NVMM)" ! cudaupload ! rrcudawrapperfilter in-place=true ! cudadownload ! fakesink