NVIDIA VPI GStreamer Plug-in -

GstVPI
GstVPI Basics
Getting Started
Evaluating GstVPI Getting the Code Building the Plugin
Examples
Gaussian Filter Undistort Harris Keypoints Detector KLT Tracker Perspective Warp Overlay Videoconvert Videoscale Convolution Separable Convolution
Performance
Nano TX1 TX2 Xavier
Contact Us

Introduction

The vpiconvolution element applies the Convolution algorithm from the NVIDIA VPI library to a video stream. In order to use this algorithm you must provide the kernel matrix to be applied.

Element properties

• backend
  

Backend to use to execute VPI algorithms. Available options:

cpu: CPU backend

cuda: CUDA backend

pva: PVA backend (Xavier only)

vic: VIC backend

Flags: readable, writable
Default: "cuda"

• kernel
  

Convolution kernel to be applied with a minimum size of 1x1 and maximum size 11x11. Usage example:

<<1.0, 0.0, -1.0>, <0.0, 0.0, 0.0>, <-1.0, 0.0, 1.0>>

Type: GstValueArray of GValues of type GstValueArray
Flags: readable, writable
Default: <<1>>

• boundary
  

How pixel values outside of the image domain should be treated. Available options:

zero: All pixels outside the image are considered 0.

clamp: Border pixels are repeated indefinitely.

Flags: readable, writable
Default: "zero"

Example

The following example pipe will help you apply convolution to the input video stream from your camera. You may modify the properties values according to the information above.

For a convolution given by an edge detection kernel of size 3x3, and a boundary condition of zero you may use:

gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! vpiupload ! vpiconvolution kernel="<<1.0, 0.0, -1.0>, <0.0, 0.0, 0.0>, <-1.0, 0.0, 1.0>>" ! vpidownload ! nvvidconv ! 'video/x-raw(memory:NVMM),format=I420' ! nvoverlaysink