Exploitation of Neural Networks for Fusion of Camera and Millimeter-Wave Radar Data

Bořek Reich

Supervisor(s): Pavel Zemčík

Brno University of Technology


Abstract: This paper introduces a multiple sensor synchronization that exploits using only off-the-shelf products. Sensor synchronization is significant for sensor fusion, an important area of research in the field of computer vision. It is known that a combination of multiple sensors can improve performance of a system and overcome the disadvantages of a single sensor, assuming sensors used are synchronized. Without proper synchronization, any attempts to benefit from sensor fusion, are useless. This study aims to explore the fusion of camera data and millimeter-wave radar data for detection and recognition purposes as well as for machine learning. This paper proposes a multiple sensor synchronization procedure that is easy to adopt for different sensors and can be used for combination of visual and non-visual sensors using only off-the-shelf products. This synchronization technique is suited for research purposes as well as some real-world applications.
Keywords: Computer Vision
Full text:
Year: 2022