# Online Adaptation of Visual Odometry Frontends with Image-Conditioned Reinforcement Learning

arXiv: https://arxiv.org/abs/2603.21785 · HTML: https://mlsanity.com/paper/2603.21785/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot navigation and UAV control, one of 15 directions in Deep reinforcement learning methods.
- Builds on 27 of its 47 references that are in the map.
- First posted on arXiv 23 Mar 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot navigation and UAV control](https://mlsanity.com/deeprl/rl-for-robot-navigation-and-uav-control/)

## Builds on (most cited first, 27 of 27)

- [Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains](https://mlsanity.com/paper/2006.10739/) (2020)
- [DUSt3R: Geometric 3D Vision Made Easy](https://mlsanity.com/paper/2312.14132/) (2023)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [SuperPoint: Self-Supervised Interest Point Detection and Description](https://mlsanity.com/paper/1712.07629/) (2017)
- [ORB-SLAM: a Versatile and Accurate Monocular SLAM System](https://mlsanity.com/paper/1502.00956/) (2015)
- [Grounding Image Matching in 3D with MASt3R](https://mlsanity.com/paper/2406.09756/) (2024)
- [ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM](https://mlsanity.com/paper/2007.11898/) (2020)
- [ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras](https://mlsanity.com/paper/1610.06475/) (2016)
- [DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras](https://mlsanity.com/paper/2108.10869/) (2021)
- [LightGlue: Local Feature Matching at Light Speed](https://mlsanity.com/paper/2306.13643/) (2023)
- [TartanAir: A Dataset to Push the Limits of Visual SLAM](https://mlsanity.com/paper/2003.14338/) (2020)
- [Direct Sparse Odometry](https://mlsanity.com/paper/1607.02565/) (2016)
- [MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors](https://mlsanity.com/paper/2412.12392/) (2024)
- [Deep Patch Visual Odometry](https://mlsanity.com/paper/2208.04726/) (2022)
- [DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networks](https://mlsanity.com/paper/1709.08429/) (2017)
- [MASt3R-SfM: a Fully-Integrated Solution for Unconstrained Structure-from-Motion](https://mlsanity.com/paper/2409.19152/) (2024)
- [VGGT-SLAM: Dense RGB SLAM Optimized on the SL(4) Manifold](https://mlsanity.com/paper/2505.12549/) (2025)
- [TartanVO: A Generalizable Learning-based VO](https://mlsanity.com/paper/2011.00359/) (2020)
- [The TUM VI Benchmark for Evaluating Visual-Inertial Odometry](https://mlsanity.com/paper/1804.06120/) (2018)
- [Deep Patch Visual SLAM](https://mlsanity.com/paper/2408.01654/) (2024)
- [TartanGround: A Large-Scale Dataset for Ground Robot Perception and Navigation](https://mlsanity.com/paper/2505.10696/) (2025)
- [VGGT-SLAM 2.0: Real-time Dense Feed-forward Scene Reconstruction](https://mlsanity.com/paper/2601.19887/) (2026)
- [Efficient Contextual Bandits with Continuous Actions](https://mlsanity.com/paper/2006.06040/) (2020)
- [SubT-MRS Dataset: Pushing SLAM Towards All-weather Environments](https://mlsanity.com/paper/2307.07607/) (2023)
- [Team CERBERUS Wins the DARPA Subterranean Challenge: Technical Overview and Lessons Learned](https://mlsanity.com/paper/2207.04914/) (2022)
- [Reinforcement Learning Meets Visual Odometry](https://mlsanity.com/paper/2407.15626/) (2024)
- [xVIO: A Range-Visual-Inertial Odometry Framework](https://mlsanity.com/paper/2010.06677/) (2020)

## Abstract (arXiv)

Visual odometry (VO) frontends are typically tuned offline by domain experts on pre-recorded datasets and then deployed with fixed hyperparameters. Yet a configuration that performs best on a benchmark is not guaranteed to remain best when texture, illumination, motion blur, sensor noise, or computational conditions change at deployment. We propose a frontend that instead adapts its parameters automatically and continuously. We formulate frontend tuning as a sequential decision-making problem and introduce an image-conditioned reinforcement-learning policy that combines a lightweight embedding of the current image with a compact set of frontend statistics. At each decision step, the policy selects the FAST detection threshold, KLT patch size, and RANSAC rejection threshold; a privileged critic provides additional context only during training. Trained on synthetic TartanAirV2 data, the policy transfers zero-shot to a monocular-inertial OpenVINS pipeline on EuRoC, TUM-VI, and UZH-FPV. On synthetic test sequences, the learned policy improves the tracking-computation trade-off over an optimized static parameter configuration. On the three real-world benchmarks, it reduces mean ATE by up to 8% and runtime by up to 57% relative to static parameters baseline. These results show that image-conditioned online adaptation can improve the accuracy-computation trade-off beyond a single configuration selected offline.