Contact Us

Computer vision

""

We are a team passionate about technology and Computer Vision methods.

The Lab's missions

To develop solutions at the cutting edge of innovation, we work on a technology watch, the implementation of new algorithms via concrete use cases and the evaluation of their results.

From the processing of images (especially satellite images) and pre-processed videos, our Computer Vision laboratory has been able to provide innovative answers to several problems such as classification, detection, ensembling, and visualization.

Our research topics

Classification

Scan an area and estimate the photovoltaic potential of the buildings in it. This is done by associating the results of a detection with the solar radiation data available on the Indrabot.

Detection

A pipeline for detecting car breakdowns. By combining advanced detection algorithms based on deep learning (YOLOv5, EfficientDet...) and more classical approaches (morphological transforms, Hough transform...)

Ensembling

Improve our performance on wild deposit detection by developing ensembling methods (AND, OR, WBF, etc.)

Visualisation

Challenge Everest, which objective was to detect dangerous positions on the ski lifts of a ski resort from video files.

NOS SOLUTIONS D'IA AU SERVICE DE LA Computer vision

Our publications

image_background

Boosting search engine capabilities of RegReview:…

RegReview is an AI solution for compliance teams, to automate regulatory monitoring and processes, which brings together several tools, the most essential of which is a search engine operating on a compiled database of custom-built regulatory sources.

The database contains ~300k documents.

2023

Read more
Article title

Labeling text clusters with keywords

We propose to explore several keyword extraction techniques to label text clusters obtained after a Text Clustering or a Topic Modeling pipeline. This work is following our previous articles about Topic Modeling and Text Clustering (here and here).

2023

Read more
image_eng

Synthetic data or how to share sensitive data…

For a period of six months, 5 students from Centrale Supélec and ESSEC worked collaboratively with Sia Partners on building a Python library to create fake - which we'll call synthetic - data.
But what's the point of creating fake data? How could it help organizations?

2023

Read more
See our publications

Contact us

*Required fields

This site is protected by reCAPTCHA and by the Google system.

Privacy Policy and Terms of Service apply.