# Geometric Consistency for Self-Supervised End-to-End Visual Odometry

arXiv: https://arxiv.org/abs/1804.03789 · HTML: https://mlsanity.com/paper/1804.03789/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Apr 2018.

## Built on by (most cited first, 11 of 11)

- [XVO: Generalized Visual Odometry via Cross-Modal Self-Training](https://mlsanity.com/paper/2309.16772/) (2023)
- [Learning Correspondence Uncertainty via Differentiable Nonlinear Least Squares](https://mlsanity.com/paper/2305.09527/) (2023)
- [Fusing Structure from Motion and Simulation-Augmented Pose Regression from Optical Flow for Challenging Indoor Environments](https://mlsanity.com/paper/2304.07250/) (2023)
- [Pose-GNN : Camera Pose Estimation System Using Graph Neural Networks](https://mlsanity.com/paper/2103.09435/) (2021)
- [On-device Self-supervised Learning of Visual Perception Tasks aboard Hardware-limited Nano-quadrotors](https://mlsanity.com/paper/2403.04071/) (2024)
- [Training on the Fly: On-device Self-supervised Learning aboard Nano-drones within 20 mW](https://mlsanity.com/paper/2408.03168/) (2024)
- [Benchmarking Visual-Inertial Deep Multimodal Fusion for Relative Pose Regression and Odometry-aided Absolute Pose Regression](https://mlsanity.com/paper/2208.00919/) (2022)
- [A high-precision self-supervised monocular visual odometry in foggy weather based on robust cycled generative adversarial networks and multi-task learning aided depth estimation](https://mlsanity.com/paper/2203.04812/) (2022)
- [USegScene: Unsupervised Learning of Depth, Optical Flow and Ego-Motion with Semantic Guidance and Coupled Networks](https://mlsanity.com/paper/2207.07469/) (2022)
- [Pose Constraints for Consistent Self-supervised Monocular Depth and Ego-motion](https://mlsanity.com/paper/2304.08916/) (2023)
- [An Attention-Based Deep Learning Architecture for Real-Time Monocular Visual Odometry: Applications to GPS-free Drone Navigation](https://mlsanity.com/paper/2404.17745/) (2024)

## Abstract (arXiv)

With the success of deep learning based approaches in tackling challenging problems in computer vision, a wide range of deep architectures have recently been proposed for the task of visual odometry (VO) estimation. Most of these proposed solutions rely on supervision, which requires the acquisition of precise ground-truth camera pose information, collected using expensive motion capture systems or high-precision IMU/GPS sensor rigs. In this work, we propose an unsupervised paradigm for deep visual odometry learning. We show that using a noisy teacher, which could be a standard VO pipeline, and by designing a loss term that enforces geometric consistency of the trajectory, we can train accurate deep models for VO that do not require ground-truth labels. We leverage geometry as a self-supervisory signal and propose "Composite Transformation Constraints (CTCs)", that automatically generate supervisory signals for training and enforce geometric consistency in the VO estimate. We also present a method of characterizing the uncertainty in VO estimates thus obtained. To evaluate our VO pipeline, we present exhaustive ablation studies that demonstrate the efficacy of end-to-end, self-supervised methodologies to train deep models for monocular VO. We show that leveraging concepts from geometry and incorporating them into the training of a recurrent neural network results in performance competitive to supervised deep VO methods.