Status: Code will be released by end of February
Paper: LEVIO: Lightweight Embedded Visual Inertial Odometry for Resource-Constrained Devices
(Accepted at IEEE Sensors Journal)
This repository will contain the official implementation of LEVIO, a lightweight embedded Visual-Inertial Odometry (VIO) system designed for resource-constrained devices such as microcontrollers and ultra low-power Systems on Chip (SoCs).
The release will include:
- An optimized implementation for the GAP9 SoC by GreenWaves Technologies
- A golden model used for algorithm development and evaluation in the accompanying paper
LEVIO is designed with:
- Minimal memory footprint
- Low computational overhead leveraging parallelization
- Real-time performance on ultra-low power embedded devices
Our approach targets applications where traditional VIO solutions are too heavy to run, enabling precise motion estimation for IoT, wearable devices, mobile robotics, and AR/VR on embedded systems.
The complete source code and documentation will be made publicly available by the end of February.