Visual-Inertial Odometry (VIO) with Spectacular AI SDK

Spectacular AI SDK fuses data from cameras and IMU sensors (accelerometer and gyroscope) and outputs an accurate 6-degree-of-freedom pose of a device. This is called Visual-Inertial Odometry (VIO) and it can be used in, among other cases, tracking (autonomous) robots and vehicles, as well as Augmented, Mixed and Virtual Reality.

Spectacular AI SDK examples

 https://github.com/SpectacularAI/sdk-examples

Supported devices

Out-of-the-box

The SDK supports a limited set of devices out-of-the-box. This means that the SDK can be used without any manual calibration, integration or parameter tuning, with these devices. If you want to test the SDK as easily as possible, we recommend buying one of these devices.

• OAK-D by Luxonis. Other stereo OAK variants with IMU sensors (e.g., OAK-D-PoE, OAK-D-CM3/4) are also supported, but not monocular systems nor OAK-D-Lite. There is an easy-to-use Python wrapper and a C++ version.
  
  
• Intel RealSense D455 and D435i (notice the "i", which is for IMU and is a must-have. D435 is not supported). Currently only supported on x86-64. Available as a C++ version.

https://github.com/SpectacularAI/sdk

The SDK is available for non-commercial use with Luxonis OAK-D device family for Python with pip install spectacularai. The C++ version for OAK-D and additionally for Intel RealSense D455 & D435i you can download from the Releases on the right.

For more alternatives (support for other devices), CPU architectures (ARM) and commercial licensing options, please contact us at https://www.spectacularai.com.