Master student at Ensae Paris (National School of Statistics and Economic Administration) and Institut Polytechnique de Paris in machine learning and statistics. Currently looking for a PhD position in computer vision and deep learning.
Previous research intern at CNRS in computer vision and deep learning.
Interested in deep learning and computer vision.
** Currently working on **
- Semantic segmentation for in situ TEM images and videos. Repo
- Generating labeled data for image segmentation (GAN and DDPM).
Contact: lucas[dot]degeorge[at]ensae[dot]fr
| lucasdegeorge[at]outlook[dot]fr
| Linkedin
Semantic segmentation for in situ TEM images. Repo
Development of semi-supervised models for segmenting nanowire images. The aim is to enable certain geometric properties (contact angle, phase length, etc.) to be measured. Under the supervision of Federico Panciera.
Unsupervised methods for multi-modal machine translation. Repo
The project is based on the works of Su et al. 2018 and Huang et al. 2020 and done under the supervision of Matthieu Futeral from INRIA.
More details available soon.
Recovering the posterior doubly-intractable distribution of a parameter using the Exchange algorithm. Repo
This project is based on the first numerical experiement of this paper. It aims at recovering the posterior doubly-intractable distribution of a parameter by
- implementing a Gibbs sampler to sample from the Ising model.
- implementing an exchange algorithm to sample from the posterior distribution. More details available here
C++ implementation of John Conway's Game of life with GUI (using Direct2D). Repo
We have implemented a simulation of the Game of Life. The simulation allows users to see the behaviour of cells in the game and to interact with the game by changing the state of cells in the grid.