Integration for DIS Example Application

Follow us on:

Share This Page

Facebook Twitter LinkedIn WhatsApp Email Copy Link

Close

Send Email Using

Gmail (Web) Thunderbird (default mail app)

Close

RidgeRun Video Stabilization Library

< >

RidgeRun Video Stabilization Library

Table of Contents [Sticky]

• Basics and Foundation
  • What is Video Stabilization
  • Video Stabilization Process using IMU
  • Stabilization Algorithms with IMU
  • Video Stabilization Process using Optical Flow
• Getting Started
  • Supported Platforms, Sensors and Backends
  • Evaluating the Library
  • Getting the Code
  • Building the Library
• API Reference
  • Library Architecture
  • Adding New Sensors
  • Adding Algorithms
  • Adding Stabilization Algorithm
  • Adding Backends
  • Adding OpticalFlow Algorithms
  • Adding Transformation Algorithms
  • API Documentation
  • Examples Guidebook
• Library Integration for IMU
  • Preparing the Video
  • Preparing the IMU Measurements
  • Measurement Integration
  • Interpolation
  • Computing the Stabilization
  • Video Undistortion
  • Example Application
• Library Integration for DIS
  • Preparing the Frames
  • Preparing the Runtime Settings
  • Example Application
• GStreamer
  • Add Support for Video Timestamps
  • Add Support for IMU Sensors
  • Video Stabilizer Element
  • Analyzer Element
  • Example Pipelines
• Calibration Process
  • Installing Calibration Tool
  • Camera Calibration
  • IMU Calibration
  • Tuning Process
• Useful Links
  • BMI160 Setup
  • ICM42688 Setup
  • Performance
  • Jetson setup
• Contact Us

Library Integration for DIS Example Application

Introduction

This example demonstrates how to stabilize a video using the Digital Image Stabilization (DIS) pipeline. It reads frames from a file or camera source using OpenCV, prepares them as RVS image objects, applies the DigitalStabilizer, and writes or displays the stabilized output.

Concept

The application implements a complete offline video stabilization pipeline. The process consists of opening an input source, configuring the stabilization parameters, preparing the input and output images, processing each frame through the DigitalStabilizer, and finally displaying or storing the stabilized output.

Instances

This application uses the following library components:

• An Optical Flow instance with Optical Flow parameters.
• A Smoothing instance with Smoothing parameters.
• A Geometric Transform instance.
• A DigitalStabilizer with DigitalStabilizer parameters.
• A HostAllocator or CudaAllocator, depending on the selected backend.

Preprocessing and Preparation

The input video is read from a file or camera source using OpenCV. Before processing begins, the application verifies that it is possible to:

• Open the input video.
• Create the output video file, if recording is enabled.

The input video properties, such as the frame size, frame rate, and codec, are obtained from the source and used to configure the output stream.

If these steps are successful, the application advances to the main processing loop.

Main Processing Loop

The main processing loop reads one frame at a time from the input source. Each frame goes through the following stages before the stabilized output is produced.

Preparing the Input and Output Images

Each captured frame is converted from the OpenCV BGR format to RGBA and wrapped as an RVS image object compatible with the selected backend.

For the OpenCV backend, the example wraps the images using the HostAllocator. For the CUDA backend, it uses persistent CudaAllocator images backed by pinned memory.

Applying the Stabilizer

Once the input and output images have been prepared, the application invokes the DigitalStabilizer.

For the OpenCV backend:

stab_ret =
    stabilizer.Apply<HostAllocatorType>(
        rvs_out,
        rvs_frame);

For the CUDA backend:

stab_ret =
    stabilizer.Apply<CudaAllocatorType>(
        rvs_out_cuda,
        rvs_frame_cuda);

The DigitalStabilizer internally executes the DIS pipeline by estimating frame-to-frame motion, smoothing the resulting motion, and applying the corresponding geometric transformation.

Processing the Output

The stabilized output is produced as an RGBA image. Before displaying or writing the frame using OpenCV, it is converted back to the BGR format.

If display mode is enabled, the stabilized frame is displayed. If recording is enabled, the stabilized frame is written to the output video file.

Cleanup

At the end of the process, the application releases its resources, prints performance statistics and summary information, and cleans up CUDA resources when the CUDA backend is used.

Back to top ↑ < >