Getting Started - Supported Platforms, Sensors and Backends

< >

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

The RidgeRun Video Stabilization Library is intended to work on GPU-accelerated systems with support for V4L2 video capture. The minimum requirements for a system to support the library are:

• Run a Linux-based Operating System.
• Have an OpenCL-capable or CUDA-capable GPU if acceleration is wanted.
• A V4L2 driver with a valid SOF/EOF monotonic timestamp (one referenced in the Linux Clock Monotonic Raw) is highly recommended.

Regarding the V4L2 driver, it is necessary to have a valid timestamp that is equal to the CLOCK_MONOTONIC_RAW or similar that can be used to compute an offset with respect to the CLOCK_MONOTONIC_RAW. For instance, if a hardware RTC exists, the offset between the RTC and the CLOCK_MONOTONIC_RAW must be known at runtime.

Tested Platforms

We have tested the RidgeRun Video Stabilization Library on the following platforms:

• x86 with Intel Iris GPU
• Qualcomm RB5
• NXP i.MX8
• NVIDIA Jetson Xavier NX

Other platforms are supported since our library is built on OpenCV, OpenCL and CUDA.

Supported Backends

An execution backend is the algorithm implementation that allows hardware acceleration. The RidgeRun Video Stabilization Library has the following supported backends:

• CPU: through OpenCV
• OpenCL: developed and optimized by RidgeRun
• CUDA: developed and optimized by RidgeRun

Supported Sensors

Currently, we have support for:

• The embedded Qualcomm RB5 Evaluation Kit IMU
• The BMI160 over I2C

The library can be extended to provide more support. You can find more information in Adding New Sensors.

Supported GStreamer video source elements

Please, check this wiki.

< >