Semi-Direct Visual Odometry Based on Monocular Depth Estimation

Among various visual sensors used in various visual odometry methods, monocular cameras are more suitable for small mobile platforms due to their low costs, light weights and low power consumptions. However, the depth of the scene cannot be measured by monocular camera so all kinds of monocular visual odometry or SLAM algorithms inevitably have the defect of scale ambiguity, which greatly limits their application in real world. To solve the above problems, a semi-direct visual odometry algorithm based on monocular depth estimation is proposed in this paper, we add a monocular depth estimation module into the semi-direct visual odometry, which provides good initial values to the depth filter. In this way, the inherent defect of scale ambiguity is overcome, and the accuracy and robustness of monocular visual odometry are improved.

[1]  Daniel Cremers,et al.  Semi-dense Visual Odometry for a Monocular Camera , 2013, 2013 IEEE International Conference on Computer Vision.

[2]  Dongbing Gu,et al.  UnDeepVO: Monocular Visual Odometry Through Unsupervised Deep Learning , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[3]  G. Klein,et al.  Parallel Tracking and Mapping for Small AR Workspaces , 2007, 2007 6th IEEE and ACM International Symposium on Mixed and Augmented Reality.

[4]  Sen Wang,et al.  DeepVO: Towards end-to-end visual odometry with deep Recurrent Convolutional Neural Networks , 2017, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[5]  J. M. M. Montiel,et al.  ORB-SLAM: A Versatile and Accurate Monocular SLAM System , 2015, IEEE Transactions on Robotics.

[6]  James R. Bergen,et al.  Visual odometry , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[7]  Tom Drummond,et al.  Faster and Better: A Machine Learning Approach to Corner Detection , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Nassir Navab,et al.  Deeper Depth Prediction with Fully Convolutional Residual Networks , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[9]  Rob Fergus,et al.  Depth Map Prediction from a Single Image using a Multi-Scale Deep Network , 2014, NIPS.

[10]  Federico Tombari,et al.  CNN-SLAM: Real-Time Dense Monocular SLAM with Learned Depth Prediction , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  Oisin Mac Aodha,et al.  Unsupervised Monocular Depth Estimation with Left-Right Consistency , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  Rob Fergus,et al.  Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-scale Convolutional Architecture , 2014, 2015 IEEE International Conference on Computer Vision (ICCV).

[13]  Noah Snavely,et al.  Unsupervised Learning of Depth and Ego-Motion from Video , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Andreas Geiger,et al.  Are we ready for autonomous driving? The KITTI vision benchmark suite , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[15]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[16]  Roland Siegwart,et al.  Kinect v2 for mobile robot navigation: Evaluation and modeling , 2015, 2015 International Conference on Advanced Robotics (ICAR).