Eduardo Arnold

Eduardo Arnold

Machine Learning Engineer



I’m a Machine Learning Engineer working on visual relocalisation at Niantic. Previously, I did my PhD at the University of Warwick, in the Intelligent Vehicles group, and was supervised by Mehrdad Dianati and Paul Jennings. My research focused on perception methods for autonomous driving, including cooperative 3D object detection, point cloud registration and sensor pose optimisation.


  • Computer Vision
  • Machine Learning
  • 3D Deep Learning
  • 3D Reconstruction
  • Visual Relocalisation


  • PhD in Machine Learning for Autonomous Driving, 2017 - 2021

    University of Warwick (UK)

  • BSc in Electrical Engineering with Honours, 2011 - 2017

    Federal University of Santa Catarina (Brazil)



Machine Learning Engineer


May 2022 – Present London, UK
  • Working on visual relocalisation;
  • Deploying systems on cloud infrastructure using K8s and ArgoCD.

PhD Research Intern


Jun 2021 – Jan 2022 London, UK
  • Lead the research and development of a novel method for relative camera pose estimation between a pair of images with focus on visual localisation.
  • Collaborated with an interdisciplinary research team.
  • Trained and deployed models on cloud infrastructure.

Data Study Group Researcher

Alan Turing Institute

Sep 2020 – Sep 2020 London, UK
Worked with a team of skilled international researchers to design and evaluate machine learning models that predict long-term wind speed statistics given historical satellite data. The results of this collaboration can be used to identify suitable locations to install wind turbines.

Teaching Assistant

University of Warwick

May 2019 – Apr 2020 Coventry, United Kingdom
  • Prepared and delivered tutorials on PyTorch and fundamentals of machine learning;
  • Prepared module assessments and marked students;
  • Created a Dockerized Jupyterhub platform for students to run their code remotely.

Research Assistant

University of Warwick

Aug 2018 – Jun 2021 Coventry, United Kingdom

The L3 Pilot European Consortium studies the impact of vehicle automation with key European OEMs. My main responsibilities were:

  • Lead the creation of a responsive online platform to support data visualisation of piloting data across the project.
  • Attend international meetings with key partners in the automotive sector to meet project requirements and report results.

Recent Publications

(2022). Map-free Visual Relocalization: Metric Pose Relative to a Single Image. ECCV 2022 - European Conference on Computer Vision.

PDF Code Dataset Preprint Video

(2021). Fast and Robust Registration of Partially Overlapping Point Clouds. 2021 IEEE Robotics and Automation Letters (RA-L).

PDF Code Dataset Preprint

(2021). Data Study Group Final Report: Greenvest Solutions. September 2020 Alan Turing Data Study Group.


(2020). Cooperative Perception for 3D Object Detection in Driving Scenarios using Infrastructure Sensors. IEEE Transactions on Intelligent Transportation Systems.

PDF Code Dataset Preprint

(2020). The L3Pilot Data Management Toolchain for a Level 3 Vehicle Automation Pilot. Electronics.



Ray Tracer from Scratch

Created a simple ray tracer renderer from scratch using C++.


Automatic traffic flow and density estimation based on computer vision.


A C++ Sudoku Solver and Generator.

Recent & Upcoming Talks