SelfVIO: Self-Supervised Deep Monocular Visual-Inertial Odometry and Depth Estimation
This addresses the data limitation issue for robotics and autonomous systems by enabling accurate ego-motion and depth estimation without calibration or labeled data, though it is incremental as it builds on existing self-supervised methods.
The paper tackles the problem of visual-inertial odometry and depth estimation without labeled data by proposing SelfVIO, a self-supervised deep learning approach that uses adversarial training and sensor fusion, achieving state-of-the-art performance in pose estimation and depth recovery on datasets like KITTI, EuRoC, and Cityscapes.
In the last decade, numerous supervised deep learning approaches requiring large amounts of labeled data have been proposed for visual-inertial odometry (VIO) and depth map estimation. To overcome the data limitation, self-supervised learning has emerged as a promising alternative, exploiting constraints such as geometric and photometric consistency in the scene. In this study, we introduce a novel self-supervised deep learning-based VIO and depth map recovery approach (SelfVIO) using adversarial training and self-adaptive visual-inertial sensor fusion. SelfVIO learns to jointly estimate 6 degrees-of-freedom (6-DoF) ego-motion and a depth map of the scene from unlabeled monocular RGB image sequences and inertial measurement unit (IMU) readings. The proposed approach is able to perform VIO without the need for IMU intrinsic parameters and/or the extrinsic calibration between the IMU and the camera. estimation and single-view depth recovery network. We provide comprehensive quantitative and qualitative evaluations of the proposed framework comparing its performance with state-of-the-art VIO, VO, and visual simultaneous localization and mapping (VSLAM) approaches on the KITTI, EuRoC and Cityscapes datasets. Detailed comparisons prove that SelfVIO outperforms state-of-the-art VIO approaches in terms of pose estimation and depth recovery, making it a promising approach among existing methods in the literature.