Cutting-edge AI expertise

Our Data Science team gathered the most state-of-the-art AI skills, applied to image processing and video analysis, in order to capitalize on existing road traffic cameras

We have strong partnerships with main AI leaders:


AI functions available has developed so for key technological bricks, which are then assembled to build robust and industrialized integrated software products.


To follow a given vehicle or object from one frame to another

Lane segmentation

To pre-select and identify different areas within an image

Shape recognition

To identify and recognize determined shaep in an image


To distribute these objects into specific categories

Face detection

To count the number of occupants while systematically blurring faces.

Quality evaluation

To identify the images most likely to guarantee the detection


To match license plates from their unique fingerprint


To blur all visible personal data and ensure privacy

Image processing and video analysis

Giving eyes to computers

By pushing further the limits of image processing, AI has enabled the disruptive emergence of the Computer Vision field.

Bringing the most from roadside cameras

These cutting-edge technologies make it possible to leverage cameras like never before, whether they are new or already installed.


Deep Learning and Machine learning

The race for performance

Thanks to the AI algorithms learning capacity, our technologies always have room for performance improvement.

Trainings on millions of images

Already trained on millions of images, our algorithms are now able to provide an extremely fine understanding capacity.


Industrial capacity for at-scale AI

Powerful and optimized calcul infrastructures

We use specific optimization methods on calculation machines to run AI at-scale and make the computing architecture always more fluid.

For a successful and customized integration

By adapting each time to the specific needs of the project, our highly-skilled Solution Architects Team makes possible a proper and smooth integration into existing systems and tools.


A strong and recognized Data Science expertise regularly contributes to the advancement of Scientific and Industrial Research on an international scale, through scientific publications :

July 8th-12th 2020, Cancun, MEXICO
L. Souquet, N. Shvai, A. Sanogo, M. Khata, A. Hasnat, A. Llanza and A. Nakib

HyperFDA: A Bi-level optimization approach to Neural Architecture Search and Hyperparameters’ Optimization via Fractal Decomposition-based Algorithm
Accepted in the GECCO - Genetic and Evolutionary Computation Conference 2020

June 24th-26th 2020, Póvoa de Varzim, PORTUGAL
A. Meicler, A.Sanogo, N.Shvai, A.Llanza, A.Hasnat, M.Khata, E.Younes, A.Khalil, Y.Lachachi and A.Nakib

Real time Automatic Urban Traffic Management Framework Based on Convolutional Neural Network under Limited Resources Constraint
Accepted in the 17th ICIAR - International Conference on Image Analysis and Recognition 2020

September 22-25, 2019, Taipei, TAIWAN
A. Nakib

Application guided Image Quality Estimation based on Classification
Accepted in 26th IEEE ICIP - International Conference on Image Processing, 2019 logo

6 Place du Colonel Bourgoin
Leonard:Paris, 75012 Paris

Subscribe to our newsletter