Video Stabilization for Embedded Systems - User Guide - Setting up the stabilizer

Video Stabilization for Embedded Systems
Video Stabilization Basics
Getting Started
Evaluating Video Stabilization Getting the code Building Video Stabilization
User Guide
Selecting the cutoff frequency Preparing the image Setting up the stabilizer Error codes
Examples
Video file stabilization
GStreamer plug-in
Installing the plug-in Example pipelines Performance
Contact Us

This page provides a basic description of the VideoStabilizer class and its usage.

Initialization

The Init method of the VideoStabilizer class initializes all the internal structures and requires the following parameters:

Width

Type: integer

Width of the frame to stabilize.

Height

Type: integer

Height of the frame to stabilize.

Framerate

Type: double

Framerate of the video being stabilized. This is the same as the sampling frequency used by stabilization filter.

Cutoff frequency

Type: double

Cutoff frequency of the low pass stabilization filter. Must be less than the framerate.

Stabilization

The StabilizeFrame method of the VideoStabilizer class is the one in charge of processing the next frame. This method only requires two parameters: the input and output Image. It is important to notice that both must be completely configured objects and the stabilizer won't assign memory for the output.

The following snippet contains a complete example of usage (assume input_image and output_image are fully configured):

  int width = 1920;
  int height = 1080;
  double framerate = 30;
  double cutoff_frequency = 0.1;
  rr::VideoStabilizer stabilizer;
  rr::RuntimeError status = stabilizer.Init(width, height, framerate, cutoff_frequency);
  status = stabilizer.StabilizeFrame(input_image, output_image);