Open to opportunities

Marcelo Eduardo Pederiva

@marceloeduardopederi

Message

Computer Vision and Autonomous Vehicles Researcher specializing in Deep Learning.

Brazil

Message

What I'm looking for

I am looking for opportunities that foster innovation in autonomous systems and promote collaboration in a dynamic team environment.

I am a dedicated Computer Vision researcher with a strong focus on sensor fusion and Deep Learning models, particularly in the context of autonomous vehicle applications. My journey in this field has been marked by a commitment to bridging advanced research with practical engineering solutions, ensuring that my work not only contributes to academic knowledge but also translates into real-world applications.

Currently, I serve as a Machine Learning Researcher and Engineer at the Centro Tecnológico de Automoción de Galicia in Spain, where I have optimized a two-sensor fusion model for real-time 3D Object Detection on Nvidia Jetson Orin. My experience includes leading projects that deploy models using real vehicle sensor data and supervising research students, showcasing my ability to guide and mentor the next generation of engineers.

With a Ph.D. in Electrical and Computer Engineering from the University of Campinas, I have developed a robust skill set in Machine Learning, Deep Learning, and Computer Vision. My research has led to several publications, and I am passionate about continuing to innovate in the field of autonomous vehicles, contributing to safer and more efficient transportation solutions.

Experience

Work history, roles, and key accomplishments

Current

Machine Learning Researcher and Engineer

Current

Centro Tecnológico de Automoción de Galicia (CTAG)

Jan 2023 - Present (2 years 5 months)

Optimized a two-sensor fusion model to perform 3D Object Detection in real-time on Nvidia Jetson Orin. Deployed models with real vehicle sensor data, including Leopard, VLP-16, VLP-32c, HDL-32, and Ouster-64. Led an Industrial Ph.D. student for an Early fusion detection model.

3D Detection Deep Learning ROS Python PyTorch TensorFlow

Current

Machine Learning Researcher

Current

Centro Tecnológico de Automoción de Galicia, CTAG

Jan 2023 - Present (2 years 5 months)

Developed a fusion model that eliminates the need for calibration matrices between sensors. Deployed a 3D object detection model for custom sensors integrated into ROS and the company’s internal system.

3D Detection ROS Deep Learning Python C C++PyTorch TensorFlow

Research Supervisor

School of Electrical and Computer Engineering, UNICAMP

Jan 2022 - Present (3 years 5 months)

Proposed Machine Learning projects for undergraduate students. Supervised undergraduate research and Final paper projects.

Machine Learning Project Management Python

Professor-Assistant

Virtual University of São Paulo (UNIVESP)

Jan 2021 - Present (4 years 5 months)

Monitored student performance and provided feedback. Supervised undergraduate students.

Performance Monitoring Feedback Education

Doctoral Researcher

University of Campinas (UNICAMP)

Mar 2020 - Present (5 years 3 months)

Optimized a three-sensor fusion model to perform in real-time on the GPU. Designed a three-sensor fusion model for 3D Object Detection and Motion estimation, implementing Cross-Attention approaches to fuse LIDAR and camera. Developed a Monocular 3D Object Detection model and led data collection and labeling efforts.

3D Detection Cross Attention LiDAR Deep Learning Data Collection Data Labeling Python PyTorch TensorFlow

Master Researcher

University of Campinas (UNICAMP)

Mar 2018 - Present (7 years 3 months)

Developed Object Detection systems for local navigation in off-road environments. Implemented Lane Line Identification for agronomy tasks. Implemented sensor fusion (GPS, IMU, odometry) for vehicle localization.

Object Detection GPS IMU Vehicle Localization Deep Learning Python

Education

Degrees, certifications, and relevant coursework

University of Campinas (UNICAMP)

Ph.D., Electrical and Computer Engineering

Grade: 4.0/4.0 GPA

Thesis: Efficient Multi-modal Fusion for 3D Object Detection and Motion Estimation. Optimized the three-sensor fusion model to perform in real-time on the GPU. Designed a three-sensor fusion model for 3D Object Detection and Motion estimation.

University of Campinas (UNICAMP)

M.Sc., Mechanical Engineering

Grade: 3.8/4.0 GPA

Dissertation: Vision-based Landmark Identification for Autonomous Navigation. Developed Object Detection systems for local navigation in off-road environments. Implemented Lane Line Identification for agronomy tasks.