DW-VIO: Deep Weighted Visual-Inertial Odometry

Visual-inertial odometry (VIO) has made significant progress in various applications. However, one of the key challenges in VIO is the efficient and robust fusion of visual and inertial measurements, particularly while mitigating the impact of sensor failures. To address this challenge, we propose a new learning-based VIO system, i. e. , DW-VIO, which is able to integrate multiple sensors and provide robust state estimations. To this end, we design a novel deep learning-based data-fusion approach that dynamically associates information from multiple sensors to predict sensor weights for optimization. Moreover, in order to improve the efficiency, we present several real-time optimization techniques including a fast patch graph constructor and an efficient GPU-accelerated multi-factor bundle adjustment layer. Experimental results show that DW-VIO outperforms most state-of-the-art (SOTA) methods on the EuRoC MAV, ETH3D-SLAM, and KITTI-360 benchmarks across various challenging sequences. Additionally, it maintains a minimum of 20 frames per second (FPS) on a single RTX 3060 GPU with high-resolution input, highlighting its efficiency.

Authors

• Guyuan Chen
• Xiyue Guo
• Xiaokun Pan
• Yujun Shen
• Guofeng Zhang 0001
• Hujun Bao
• Zhaopeng Cui

Keywords

• Performance evaluation
• Visualization
• Graphics processing units
• Robot sensing systems
• Real-time systems
• Mobile handsets
• Odometry
• State estimation
• Optimization
• Intelligent robots
• Visual-inertial Odometry
• Benchmark
• Frames Per Second
• Bundle Adjustment
• High-resolution Input
• Image Features
• Visual Information
• Prior Information
• Visual Features
• Learning-based Methods
• Optimal Weight
• Jacobian Matrix
• Inertial Measurement Unit
• Optical Flow
• Correlated Features
• Hessian Matrix
• Feature Matching
• Deep Learning-based Methods
• Optimal Window
• Camera Frame
• Accurate State Estimation
• Inertial Measurement Unit Data
• Tangent Space
• Factor Graph
• Normal Equations
• Flashlight
• Neural Network
• State Of The Art Methods
• Single Measurement
• Dynamic Optimization