Self-Supervised Geometry-Guided Initialization for Robust Monocular Visual Odometry
This work addresses robustness issues in autonomous systems by enhancing a learning-based SLAM method, though it is incremental as it builds on existing dense SLAM frameworks.
The paper tackled the problem of monocular visual odometry struggling with large motions and object dynamics by using self-supervised priors from a pre-trained depth estimator to initialize dense bundle adjustment, resulting in significant improvements on KITTI and DDAD benchmarks.
Monocular visual odometry is a key technology in various autonomous systems. Traditional feature-based methods suffer from failures due to poor lighting, insufficient texture, and large motions. In contrast, recent learning-based dense SLAM methods exploit iterative dense bundle adjustment to address such failure cases, and achieve robust and accurate localization in a wide variety of real environments, without depending on domain-specific supervision. However, despite its potential, the methods still struggle with scenarios involving large motion and object dynamics. In this study, we diagnose key weaknesses in a popular learning-based dense SLAM model (DROID-SLAM) by analyzing major failure cases on outdoor benchmarks and exposing various shortcomings of its optimization process. We then propose the use of self-supervised priors leveraging a frozen large-scale pre-trained monocular depth estimator to initialize the dense bundle adjustment process, leading to robust visual odometry without the need to fine-tune the SLAM backbone. Despite its simplicity, the proposed method demonstrates significant improvements on KITTI odometry, as well as the challenging DDAD benchmark.