Paul Marchwica, P.Eng
About Paul Marchwica, P.Eng
Paul Marchwica, P.Eng, is a Computer Vision Developer on the Innovation Team at Sportlogiq, specializing in image registration and camera calibration. He holds a Master of Applied Science in Mechanical Engineering from the University of Windsor and has extensive experience in developing real-time libraries for video segmentation using traditional and deep learning techniques.
Work at Sportlogiq
Paul Marchwica serves as a Computer Vision Developer on the Innovation Team at Sportlogiq, a role he has held since 2018. In this position, he focuses on enhancing data gathering speed and quality in American football through automation. His work involves developing scalable, real-time libraries for video segmentation and camera calibration, utilizing both traditional and deep learning techniques. He employs tools such as Keras, TensorFlow, Python, OpenCV, Docker, and Slurm to achieve these objectives.
Education and Expertise
Paul Marchwica holds a Master of Applied Science (MASc) in Mechanical Engineering from the University of Windsor, where he studied from 2010 to 2012. He also earned a Bachelor of Applied Science (BASc) in Systems Design Engineering from the University of Waterloo, completing his studies from 2004 to 2009. His educational background provides a strong foundation for his expertise in computer vision, image registration, and camera calibration.
Background
Paul Marchwica has a diverse professional background in engineering and programming. He began his career with co-op positions at Ford Motor Company and Atlantis Cyberspace, Inc. He later worked as a C++ Programmer at Harris Communications and as a Software Developer at BlackBerry. He also served as a Senior .NET Developer at Senstar before joining Sportlogiq. Additionally, he briefly worked as an Engineering Instructor at Johns Hopkins University’s Center for Talented Youth.
Research Contributions
Paul Marchwica actively collaborates on research articles in the field of computer vision. His work includes contributions to studies on shape-matching neural networks and the identification of race and sex bias in media broadcasting. He co-authored a paper titled 'Self-Supervised Shape Alignment for Sports Field Registration,' which was presented at WACV 2022, and another paper on 'Unsupervised Domain Adaptation in Person re-ID via k-Reciprocal Clustering and Large-Scale Heterogeneous Environment Synthesis,' presented at WACV 2020.
Technical Skills
Paul Marchwica possesses a range of technical skills that support his work in computer vision. He is proficient in programming languages and frameworks such as Python and C++, and utilizes machine learning libraries like Keras and TensorFlow. His experience with OpenCV aids in image processing tasks, while Docker and Slurm facilitate efficient project management and deployment. These skills enable him to develop innovative solutions in the field of computer vision.