Portfolio risk analysis and data science for financial BI at the ICA
I participated on a project to study the measurement of paramecium ciliary beating from video.
The goal of my research is to provide a metrological study comparing well-known algorithms to compute the optical flow of video sequences. In particular, we are interested by videos of plants motion (more specifically single trees) which represent particularly complex cases for image processing techniques. This work is part of a larger research project studying the bio-mechanics of plants. Here I focus on developing an experimental method to study plant dynamics and structure using video data.
Through this project, I have coded the CR matlab toolbox containing a set of algorithms that allows the extraction of motion data out of video sequences. This toolbox proposes a user friendly interface designed to be used by scientists who are not specialized in image processing.
This post-doc is founded by the ANR research project Chêne-Roseau.
- Acquisition and reproduction of real motion from videos – project page
- Real-time simulation of the tree response to wind load – project page
- My thesis (english, ~ 5Mb, this version is still waiting for a few corrections)
- Attestation de réussite au diplôme (french)
- Procès verbal de ma soutenance et rapports du mémoire de thèse (french and one thesis report in English)