Improving CNNs Towards Real-Time Multi-modal Object Detection in Remote Sensing Imagery
Deep-learning object detection methods that are designed for computer vision applications tend to under-perform when applied to remote sensing data. This is because, contrary to computer vision, in remote-sensing targets can be very small, occupying only a few pixels in the entire image, and exhibit arbitrary perspective transformations. Detection performance can improve by fusing data from multiple remote sensing modalities, including RGB, IR, hyper-spectral, multi-spectral, synthetic aperture radar, and LiDAR, to name a few.
In this talk, I will present YOLOrs: a new convolutional neural network, specifically designed for real-time object detection in multi-modal remote sensing imagery. YOLOrs can detect objects at multiple scales, as well as predict target orientations. In addition, YOLOrs introduces a novel mid-level fusion architecture that enables its application to multi-modal aerial imagery. Our experiments corroborate the efficiency of YOLOrs compared to contemporary alternatives.
About the Speaker
Dr. Panos P. Markopoulos received the Engineering Diploma (5-year) and M.Sc. degrees in Electronic and Computer Engineering from the Technical University of Crete, Greece, and the Ph.D degree in Electrical Engineering from The State University of New York at Buffalo, USA.
Since 2015, he has been an Assistant Professor of Electrical Engineering at the Rochester Institute of Technology, Rochester NY, USA, where he directs the Machine Learning Optimization and Signal Processing Laboratory (MILOS LAB) and is a core faculty of the RIT Center for Human-aware Artificial Intelligence (CHAI). In the Summers of 2018 and 2020, he served as Visiting Research Faculty at the U.S. Air Force Research Laboratory. His research is in the areas of machine learning, signal processing, and data analysis, with a current focus on robust analysis of tensor data. In these areas, he has co-authored more than 50 peer-reviewed journal and conference articles.
Dr. Markopoulos’s research has been supported with multiple grants from the U.S. National Science Foundation, US. Air Force Office of Scientific Research, U.S. Air Force Research Lab, the U.S. Government, as well as industry partners. He is a member of IEEE Signal Processing, Computer, and Communication Societies, with high service activity including, most recently, the organization of the IEEE International Workshop on Machine Learning for Signal Processing (IEEE MLSP 2019).