Learning stochastic geometry models and Convolutional Neural Networks. Application to multiple object detection in aerospatial data sets
Josiane Zerubia
INRIA, France
ABSTRACT
Convolutional neural networks (CNN) have shown great results for object-detection tasks by learning texture and pattern-extraction filters. However, object-level interactions are harder to grasp without increasing the complexity of the architectures. On the other hand, Point Process models propose to solve the detection of the configuration of objects as a whole, allowing the factoring in of the image data and the objects prior interactions. In this talk, we propose combining the information extracted by a CNN with priors on objects within a Markov Marked Point Process framework. We also propose a method to learn the parameters of this Energy-Based Model. We apply this model to the detection of small vehicles in optical satellite imagery, where the image information needs to be complemented with object interaction priors because of noise and small object sizes. This is a joint work with my former PhD student (Jules Mabon) in collaboration with Airbus DS (Mathias Ortner).
SHORT BIO
Josiane Zerubia has been a permanent research scientist at INRIA since 1989 and Director of Research since July 1995 (DR Exceptional Class since 2023; DR 1st Class from 2002 to 2022). She was head of the PASTIS remote sensing laboratory (INRIA Sophia-Antipolis) from mid-1995 to 1997 and of the ARIANA research group (INRIA/CNRS/University of Nice), which worked on inverse problems in remote sensing and biological imaging, from 1998 to 2011. From 2012 to 2016, she was head of AYIN research group (INRIA-SAM) dedicated to models of spatio-temporal structure for high-resolution image processing with a focus on remote sensing and skincare imaging. She is head of AYANA exploratory research group since 2020. AYANA is an interdisciplinary project using knowledge in stochastic modeling, image processing, artificial intelligence, remote sensing and embedded electronics/computing. She was professor (PR1) at SUPAERO (ISAE) in Toulouse from 1999 to 2020. She received a Doctor Honoris Causa degree from the University of Szeged in Hungary in 2020, and 3 times the Excellence Award from University of Nice (now UCA) in 2020, 2019 and 2016. She supervised or co-supervised 63 Master students, 37 PhDs and 27 post-docs. She was external examinator for PhD degrees at Purdue Univ. (West-Lafayette, USA), Heriot Watt Univ. (Edinburgh, GB), Univ. of Iceland (Reyljavik, Iceland), University of Lisbon (Portugal), Univ. of Manouba (Tunis, Tunisia), Sup Telecom (Tunis, Tunisia), University of Rabat (Morocco), and for more than 30 PhDs in France including one at the University of the French West Indies.
Before that, she was with the Signal and Image Processing Institute of the University of Southern California (USC) in Los-Angeles as a postdoc (1988-1989). She also worked as a researcher for the LASSY (University of Nice/CNRS) from 1984 to 1988 and in the Research Laboratory of Hewlett Packard in France and in Palo-Alto (CA) from 1982 to 1984. She received the MSc degree from the Department of Electrical Engineering at ENSIEG, INP Grenoble, France in 1981, the Doctor of Engineering degree, her PhD and her ‘Habilitation’, in 1986, 1988, and 1994 respectively, all from the University of Nice, France.
She is a Fellow of the IEEE (2003- ), the EURASIP (2019- ) and the IAPR (2020-), and IEEE SP Society Distinguished Lecturer (2016-2017).
Her main research interest is in image processing and remote sensing using probabilistic models. She also works on parameter estimation, statistical learning, optimization techniques, and artificial intelligence.