Visual Odometry Revisited: What Should Be Learnt?
This work addresses scale-drift and performance limitations in monocular VO systems for robotics and autonomous driving applications, representing an incremental improvement by combining existing techniques.
The authors tackled the problem of monocular visual odometry (VO) by integrating deep learning with geometry-based methods, resulting in a system (DF-VO) that outperforms both pure deep learning and geometry-based approaches and eliminates scale-drift issues, as demonstrated on the KITTI dataset.
In this work we present a monocular visual odometry (VO) algorithm which leverages geometry-based methods and deep learning. Most existing VO/SLAM systems with superior performance are based on geometry and have to be carefully designed for different application scenarios. Moreover, most monocular systems suffer from scale-drift issue.Some recent deep learning works learn VO in an end-to-end manner but the performance of these deep systems is still not comparable to geometry-based methods. In this work, we revisit the basics of VO and explore the right way for integrating deep learning with epipolar geometry and Perspective-n-Point (PnP) method. Specifically, we train two convolutional neural networks (CNNs) for estimating single-view depths and two-view optical flows as intermediate outputs. With the deep predictions, we design a simple but robust frame-to-frame VO algorithm (DF-VO) which outperforms pure deep learning-based and geometry-based methods. More importantly, our system does not suffer from the scale-drift issue being aided by a scale consistent single-view depth CNN. Extensive experiments on KITTI dataset shows the robustness of our system and a detailed ablation study shows the effect of different factors in our system.