Self-Supervised Geometry-Guided Initialization for Robust Monocular Visual Odometry

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. The project page: https://toyotafrc.github.io/SGInit-Proj/

Authors

• Takayuki Kanai
• Igor Vasiljevic
• Vitor Guizilini
• Kazuhiro Shintani

Keywords

• Bundle adjustment
• Training
• Simultaneous localization and mapping
• Accuracy
• Estimation
• Self-supervised learning
• Benchmark testing
• Trajectory
• Optimization
• Visual odometry
• Monocular Visual Odometry
• Failure Cases
• Depth Estimation
• Monocular Depth Estimation
• Neural Network
• Point Cloud
• Nonlinear Programming
• Depth Map
• Learning Module
• Optical Flow
• Pose Estimation
• Standard Benchmark
• Challenging Scenarios
• Camera Model
• Camera Pose
• State Of The Art Methods
• Depth Prediction
• Accurate Depth
• Trajectory Estimation
• Dense Depth
• Camera Intrinsics
• Self-supervised Training
• Ground Truth Trajectory