Mohammad Hossein Bamorovat Abadi

About Me

Mohammad is a highly skilled and accomplished individual with a diverse background in robotics, computer vision, and machine learning. With a Ph.D. in Computer Science from the University of Hertfordshire, he has conducted extensive research in the field of human activity detection in ambient assisted living scenarios. Under the supervision of Prof. Farshid Amirabdollahian, he has excelled in creating datasets for indoor human action recognition and implementing deep learning models for visual human action recognition. Mohammad’s expertise extends to robotic software engineering, where he has contributed significantly to projects at Conigital, a leading autonomous car company. His responsibilities included driving monitoring system development, vision fusion, lidar camera fusion, and designing sensor suites. Also, he developed deep learning models using vision, point cloud, and sensor fusion techniques for object detection at Conigital. With a strong command of programming languages such as C++, Python, and C, he has also demonstrated proficiency in using software tools and platforms such as MATLAB, ROS, ROS2, OpenCV, PyTorch, Keras, TensorFlow, and LaTex. Mohammad’s dedication to his field is evident through his contributions as an R&D manager at GEIT.Co, where he led research groups and played a crucial role in designing and implementing innovative industrial and hospital robots. His exceptional leadership skills and ability to secure funded projects highlight his talent for writing proposals. Additionally, Mohammad’s teaching experience as a robotic and programming lecturer at various educational institutions in the UK and Iran underscores his commitment to knowledge sharing. With a strong foundation in robotics, computer vision and machine learning, Mohammad’s expertise and accomplishments make him a highly capable and sought-after professional.

Skills

Topics:
- Robotics, Computer Vision, Machine/Deep Learning, Vision Navigation
Programming Languages:
- C++, Python, C, MATLAB
Software Libraries and Tools:
- ROS, ROS2, RVIZ, Gazebo, OpenCV, PyTorch, SciKit-Learn, Keras, TensorFlow, Docker, AWS, GPU/CPU Clusters
IoT:
- RASPBERRY PI, ARDUINO, NVIDIA (Jetson Nano, Jetson Xavier)

Professional Experience

Robotics Software Engineer (Perception Team) - Conigital

Jan 2022 - Present

Development of a driver monitoring system.
Implementation of real-time multi-camera fusion to create panoramic views.
Calibration of multi-sensor setups (camera, lidar, radar).
Design and development of 2D/3D visual object detection models.
Advancement in lidar-camera sensor fusion for enhanced object detection.
Creation of new AV sensor suites.
Retraining and validation of object detection models with various AV datasets.

Ph.D. Research Candidate - University of Hertfordshire

Feb 2019 - Dec 2023

Conducted comprehensive research on visual datasets for Human Action Recognition (HAR).
Generated a multiview HAR dataset tailored for human-robot interaction.
Evaluated deep learning models for HAR, such as C3D, R3D, and Slow-Fast.
Developed multiview C3D models to utilize the Robot House Multiview dataset (RHM).

Visiting Lecturer - Department of Computer Science, University of Hertfordshire

Oct 2019 - Feb 2022

Machine Learning and Neural Computing
Advanced Computer Science Masters Project
Legal and Ethical Practice Exercise
Human Dimensions of Computing
Master/Bachelaor Student final project supervisor

R&D Manager - GEIT.Co

Apr 2009 - Nov 2018

Directed research in developing robotics and computer vision solutions.
Implemented industrial and hospital robots.
Developed a ROS node for SOSCO's Mobile Robot for enhanced system integration.
Organized National Student RoboCup Competitions in Iran.

Publications

Mohammad Hossein Bamorovat Abadi, Mohamad Reza Shahabian Alashti, Patrick Holthaus, Catherine Menon, and Farshid Amirabdollahian. “RHM: Robot House Multi- view Human Activity Recognition Dataset.” ACHI 2023, Venice, Italy, IARIA.
Publication Link
Mohamad Reza Shahabian Alashti, Mohammad Hossein Bamorovat Abadi, Patrick Holthaus, Catherine Menon, and Farshid Amirabdollahian. “RHM-HAR-SK: A Multi- view Dataset with Skeleton Data for Ambient Assisted Living Research.” ACHI 2023, Venice, Italy, IARIA.
Publication Link
Mohamad Reza Shahabian Alashti, Mohammad Hossein Bamorovat Abadi, Patrick Holthaus, Catherine Menon, and Farshid Amirabdollahian. “Lightweight human activity recognition for ambient assisted living.” ACHI 2023, Venice, Italy, IARIA.
Publication Link
Mohammad Hossein Bamorovat Abadi, Mohamad Reza Shahabian Alashti, Patrick Holthaus, Catherine Menon, and Farshid Amirabdollahian. “Affordable robot mapping using omnidirectional vision.” 4th UK-RAS Conference: Robotics at Home (UKRAS21), 29–30. Hatfield, UK, 2021.
Publication Link
Mohammad Hossein Bamorovat Abadi, Mohamad Reza Shahabian Alashti, Patrick Holthaus, Catherine Menon, and Farshid Amirabdollahian. “Robot house human activity recognition dataset.” 4th UK-RAS Conference: Robotics at Home (UKRAS21), 19–20. Hatfield, UK, 2021.
Publication Link
Mohamad Reza Shahabian Alashti, Mohammad Hossein Bamorovat Abadi, Patrick Holthaus, Catherine Menon, and Farshid Amirabdollahian. “Human activity recogni- tion in robocup@home: inspiration from online benchmarks.” 4th UK-RAS Conference: Robotics at Home (UKRAS21), 27–28. Hatfield, UK, 2021.
Publication Link
Mohammad Hossein Bamorovat Abadi, Mohammad Reza Asghari Oskoei, Ahmad Fakharian, “Side Sonar Vision Applied to Omni-directional Images to Navigate Mobile Robots.”, 5th Iranian joint congress on Fuzzy and Intelligent systems- cfis2017, IEEE, 2017.
Publication Link
Mohammad Hossein Bamorovat Abadi, Mohammad Reza Asghari Oskoei, “Effects of Mirrors in Mobile Robot Navigation Based on Omnidirectional Vision.” 8th Interna- tional Conference, ICIRA 2015, Portsmouth, UK, August 24-27, 2015.
Publication Link
Mohammad Hossein Bamorovat Abadi, Mohammad Reza Asghari Oskoei, Ahmad Fakharian, “Mobile robot navigation using sonar vision algorithm applied to omnidirec- tional vision.” AI & Robotics (IRANOPEN), 2015,IEEE, 1–6, 2015.
Publication Link

Education

University of Hertfordshire, Hatfield, UK

Ph.D. Student, Computer Science, Dec 2023

Research Topic: Human activity detection in ambient assisted living scenarios.
Principal Supervisor: Prof. Farshid Amirabdollahian.
Second Supervisors: Dr. Catherine Menon & Dr. Patrick Holthaus.
Main Activities included creating datasets for indoor human action recognition, implementing deep learning for visual human action recognition, and developing human tracking visual systems with Fetch Robot.

Islamic Azad University, Qazvin Branch, Qazvin, Iran

M.S., Mechatronics Engineering, 2015

Thesis Topic: Implementation of Mobile Robot Navigation using Omnidirectional Vision.
Supervisor: Mohammadreza Asghari Oskoei, Assistant Professor.
Advisors: Alireza Mohammadi Shahri & Ahmad Fakharian, Assistant Professors.
Courses included Mechatronics I & II, Advanced Robotics, and Robot Vision & Navigation.

Islamic Azad University, Sabzevar Branch, Sabzevar, Iran

B.S., Electrical & Electronics Engineering, 2009

Thesis Topic: Design and Implementation of Advance Path Following Robot.
Supervisor: Freidoon Javashi, M.S.
Relevant Courses: Image Processing, Neural Networks, Microprocessors, Control Systems.