# CNN-based Ego-Motion Estimation for Fast MAV Maneuvers

arXiv: https://arxiv.org/abs/2101.01841 · HTML: https://mlsanity.com/paper/2101.01841/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 43 references that are in the map.
- First posted on arXiv 6 Jan 2021.

## Builds on (most cited first, 24 of 24)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Identity Mappings in Deep Residual Networks](https://mlsanity.com/paper/1603.05027/) (2016)
- [Spatial Transformer Networks](https://mlsanity.com/paper/1506.02025/) (2015)
- [ORB-SLAM: a Versatile and Accurate Monocular SLAM System](https://mlsanity.com/paper/1502.00956/) (2015)
- [PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume](https://mlsanity.com/paper/1709.02371/) (2017)
- [ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM](https://mlsanity.com/paper/2007.11898/) (2020)
- [Unsupervised Learning of Depth and Ego-Motion from Video](https://mlsanity.com/paper/1704.07813/) (2017)
- [VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator](https://mlsanity.com/paper/1708.03852/) (2017)
- [Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Video](https://mlsanity.com/paper/1908.10553/) (2019)
- [Competitive Collaboration: Joint Unsupervised Learning of Depth, Camera Motion, Optical Flow and Motion Segmentation](https://mlsanity.com/paper/1805.09806/) (2018)
- [Self-supervised Learning with Geometric Constraints in Monocular Video: Connecting Flow, Depth, and Camera](https://mlsanity.com/paper/1907.05820/) (2019)
- [UnDeepVO: Monocular Visual Odometry through Unsupervised Deep Learning](https://mlsanity.com/paper/1709.06841/) (2017)
- [Relative Camera Pose Estimation Using Convolutional Neural Networks](https://mlsanity.com/paper/1702.01381/) (2017)
- [Scale-Invariant Convolutional Neural Networks](https://mlsanity.com/paper/1411.6369/) (2014)
- [AlphaPilot: Autonomous Drone Racing](https://mlsanity.com/paper/2005.12813/) (2020)
- [Inverse Compositional Spatial Transformer Networks](https://mlsanity.com/paper/1612.03897/) (2016)
- [Robust Stereo Visual Inertial Odometry for Fast Autonomous Flight](https://mlsanity.com/paper/1712.00036/) (2017)
- [Learning scale-variant and scale-invariant features for deep image classification](https://mlsanity.com/paper/1602.01255/) (2016)
- [Direct Visual-Inertial Ego-Motion Estimation via Iterated Extended Kalman Filter](https://mlsanity.com/paper/2001.05215/) (2020)
- [PRGFlow: Benchmarking SWAP-Aware Unified Deep Visual Inertial Odometry](https://mlsanity.com/paper/2006.06753/) (2020)
- [Improved State Estimation in Quadrotor MAVs: A Novel Drift-Free Velocity Estimator](https://mlsanity.com/paper/1509.03388/) (2015)

## Built on by (most cited first, 3 of 3)

- [CUAHN-VIO: Content-and-Uncertainty-Aware Homography Network for Visual-Inertial Odometry](https://mlsanity.com/paper/2208.13935/) (2022)
- [Self-Supervised Monocular Visual Drone Model Identification through Improved Occlusion Handling](https://mlsanity.com/paper/2504.21695/) (2025)
- [OpenVO: Open-World Visual Odometry with Temporal Dynamics Awareness](https://mlsanity.com/paper/2602.19035/) (2026)

## Abstract (arXiv)

In the field of visual ego-motion estimation for Micro Air Vehicles (MAVs), fast maneuvers stay challenging mainly because of the big visual disparity and motion blur. In the pursuit of higher robustness, we study convolutional neural networks (CNNs) that predict the relative pose between subsequent images from a fast-moving monocular camera facing a planar scene. Aided by the Inertial Measurement Unit (IMU), we mainly focus on translational motion. The networks we study have similar small model sizes (around 1.35MB) and high inference speeds (around 10 milliseconds on a mobile GPU). Images for training and testing have realistic motion blur. Departing from a network framework that iteratively warps the first image to match the second with cascaded network blocks, we study different network architectures and training strategies. Simulated datasets and a self-collected MAV flight dataset are used for evaluation. The proposed setup shows better accuracy over existing networks and traditional feature-point-based methods during fast maneuvers. Moreover, self-supervised learning outperforms supervised learning. Videos and open-sourced code are available at https://github.com/tudelft/PoseNet_Planar