Library Integration for IMU

< >

RidgeRun Video Stabilization Library

Table of Contents [Sticky]

• Basics and Foundation
  • What is Video Stabilization
  • Video Stabilization Process using IMU
  • Stabilization Algorithms with IMU
• 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
  • 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
• GStreamer
  • Add Support for Video Timestamps
  • Add Support for IMU Sensors
  • Video Stabilizer Element
  • Example Pipelines
• Useful Links
  • BMI160 Setup
  • Performance
• Contact Us

This section will cover constructing a video stabilizer that uses IMU data for movement estimation.

1. Preparing the Video: shows details of how to construct the Image objects regarding the format and timestamps. Each video frame must have a proper format and timestamp with the same reference as the IMU measurements and should be expressed in microseconds.
2. Preparing the IMU Measurements: shows details of how to construct the SensorPayload objects regarding the format and timestamps. Each video frame must have a proper format and timestamp with the same reference as the IMU measurements and should be expressed in microseconds. The angles must be in radians/second.
3. Measurement Integration: the vector of measurements (SensorPayload) must be integrated and transformed into softened quaternions using one of the Integration algorithms.
4. Interpolation: the integrated quaternions must be interpolated and aligned to the timestamp of each frame such that the angle can match the actual position of the camera when the image was captured. In case the image does not have a timestamp, the timestamps must be aligned accordingly before this step.
5. Computing the Stabilization: once the rotations are known for the image of interest, the stabilization estimates the distortion the video has suffered throughout the entire movement history. For that reason, we need to apply one stabilization algorithm to determine the distortion of the image.
6. Video Undistortion: with the distortion computed, we can undistort the image using any available execution backends.

We also provide an Example Application detailing this process.

< >