profile image

FOZAME ENDEZOUMOU Armand Bryan

M2 AI Student

Contact Me

About Me

I am currently a Cameroonian Master's student at Aivancity in Cachan, studying artificial intelligence, data science. This is a fascinating and vast field, and I have decided to focus more on computer vision, which has the potential to revolutionize areas like medicine and new technologies. Naturally, I have diverse skills as I enjoy research in various fields. Building ethical models simulating the human visual and nervous system to realize innovative projects and also being able to help cure eye diseases, cardiovascular conditions, and diabetes for my loved ones is a powerful motivation for me.

I am currently focused on research and work in 2D and 3D object segmentation and detection, scene recognition, medical AI. Additionally, I am involved in implementing efficient and optimized algorithms. I also delve into NLP as it allows for the development of recurrent models. To create a robust vision model, it is crucial to consider not only the visual aspect but also the recurrent and temporal aspects.

I am actively seeking a research internship in these fields.

Projects/Research [full project on my github]


project name

Violence Detection on Video

In the academic context with Dr Yasser Almehio, Deeplearning teacher, we created a model capable of classifying violent and non-violent videos. The model should be able to capture spatial and temporal information while manipulating the multidimensional nature of videos (4D) Such a neural model would be very useful for police or security forces.

Fozame Endezoumou, Harold Geumtchengh

Find out more

project name

Winter Project 2024 (Super Resolution with ESRGAN)

I am developing a Super Resolution model based on GANs, specifically ERSGAN, from scratch, which I am enhancing to increase computational speed.

It's a Personal Project

Find out more

project name

Segmentation and count of Micro organism

In this project, we conducted semantic segmentation and performed a microorganism counting in an image or a video using the YOLOV8 model.

Fozame Endezoumou, Super Aymard, Asma Ghamacha, Dave Jaydutt, Ramses Appesses

Find out more

project name

Analysis on route information failure in IP core networks by NFV-based test environment | Result(train = 94% and test = 92%)

We have developed an LSTM model to classify each state of the network over time in order to predict the state of the network's core part 10 minutes in advance.

Fozame Endezoumou, Tagueka Arnold Jordan, Tasnim Naji, Chekir Yosra, Drine Abdessalem

Find out more

project name

AI/ML for 5G-Energy Consumption Modelling by ITU AI/ML in 5G Challenge

In this challenge organized by the International Telecommunication Union, we developed a forecasting model to estimate the energy consumption by 5G over the upcoming days. My team and I were awarded the silver medal.

Fozame Endezoumou, Tagueka Arnold Jordan, Christian Nebot, Damos Ayobo