Samuel Cerezo

Electronic Engineer · PhD in Systems Engineering and Computer Science

Portrait of Samuel Cerezo

Education

Universidad de Zaragoza
Ph.D. in Systems Engineering and Computer Science, advised by Javier Civera.
Supported by competitive research FPI Grant.
2021 - 2026
Universidad Nacional del Comahue
Electronic Engineering, advised by Matías Córdoba.
Supported by YPF scholarship ($5000/y for 5 years).
GPA: 8.69 / 10.
2014 - 2020

Languages

Spanish
Native
English
Professional working proficiency
Italian
Professional working proficiency

Employment

Automation Engineer · ACelli
Developing and validating AGV control, navigation and safety functions using Siemens PLCs, CAN/CANopen and industrial sensor systems.
Jan 2026 - Present
Lucca, Italy
Doctoral Researcher · Universidad de Zaragoza
Conducted research on visual-inertial state estimation, odometry, 3D reconstruction and SLAM, leading to a PhD in Systems Engineering and Computer Science.
2021 - 2026
Zaragoza, Spain
Research Intern · KUKA
Worked on Gaussian-based 3D reconstruction using an Intel RealSense camera and a KUKA LBR iisy cobot. Developed ROS 2 tools for data recording, synchronization and alignment.
Apr 2024 - Mar 2025
Augsburg, Germany
Junior Engineer · Hydroner
Contributed to the development and deployment of wireless communication systems for IoT-based monitoring in oilfield environments.
Jun 2020 - Feb 2021
Neuquén, Argentina
Junior Automation Engineer · Matra SRL
Designed and implemented automated control systems for clients in the oil industry using Allen-Bradley and Siemens PLCs.
2019 - 2020
Neuquén, Argentina
Assistant Professor · Universidad Nacional del Comahue
Taught undergraduate Control Systems I, prepared lectures and guided laboratory sessions covering theoretical and practical concepts.
2018 - 2020
Neuquén, Argentina

Publications

SLAM&Render: A Benchmark for the Intersection Between Neural Rendering, Gaussian Splatting and SLAM
S. Cerezo, G. Meli, T. Berriel, K. Safronov, J. Civera
A novel dataset designed to benchmark methods in the intersection between SLAM and novel view rendering. It consists of 40 sequences with synchronized RGB, depth, IMU, robot kinematic data, and ground-truth pose streams.
IEEE IROS, 2026
DefVINS: Visual-Inertial Odometry for Deformable Scenes
S. Cerezo, J. Civera
A new deformable visual-inertial odometry framework that separates a rigid, IMU-anchored state from a non-rigid warp represented by an embedded deformation graph.
Submitted to IEEE ICRA 2027
An Efficient Closed-Form Solution to Full Visual-Inertial State Initialization
S. Cerezo, S. Lee, J. Civera
A new analytical solution that is easy to implement and robust at initialization, thanks to the small-rotation and constant-velocity approximations, which simplify the problem while preserving the essential coupling between motion and inertial measurements.
IEEE RA-L, 2026
GNSS-Inertial State Initialization Using Inter-Epoch Baseline Residuals
S. Cerezo, J. Civera
A novel GNSS-inertial initialization strategy that delays the use of global GNSS measurements until sufficient information is available to accurately estimate the state of a sensorized device. A criterion based on the evolution of the Hessian matrix singular values is introduced.
IEEE RA-L, 2025
Camera Motion Estimation from RGB-D-inertial Scene Flow
S. Cerezo, J. Civera
Estimates camera motion and IMU state in a rigid 3D environment, with the flexibility to operate as a multiframe optimization or to marginalize older data.
IEEE/CVF CVPR Workshop, 2024