Self-Supervised Deep Pose Corrections for Robust Visual Odometry

We present a self-supervised deep pose correction (DPC) network that applies pose corrections to a visual odometry estimator to improve its accuracy. Instead of regressing inter-frame pose changes directly, we build on prior work that uses data-driven learning to regress pose corrections that account for systematic errors due to violations of modelling assumptions. Our self-supervised formulation removes any requirement for six-degrees-of-freedom ground truth and, in contrast to expectations, often improves overall navigation accuracy compared to a supervised approach. Through extensive experiments, we show that our self-supervised DPC network can significantly enhance the performance of classical monocular and stereo odometry estimators and substantially out-performs state-of-the-art learning-only approaches.

Authors

• Brandon Wagstaff
• Valentin Peretroukhin
• Jonathan Kelly

Keywords

• Image reconstruction
• Cameras
• Training
• Lighting
• Pipelines
• Robustness
• Visual odometry
• Pose Correction
• Pose Changes
• Error Of The Mean
• Learning Models
• Convolutional Neural Network
• Deep Neural Network
• Target Image
• Test Sequences
• Depth Map
• Source Images
• ReLU Activation
• Optical Flow
• Pose Estimation
• Lie Algebra
• Valid Sequences
• Dynamic Objects
• Classical Estimation
• Loop Closure
• Warped Image
• Monocular Images
• Ground Truth Pose
• Warping Function
• Training Epochs
• Loss Function
• Camera Intrinsics
• Pixel Intensity
• Training Sequences
• Depth Prediction