Denis Mbey Akola

Graduate student

MSc student

About me

I am graduate with a Master’s degree (obtained in 2023) in Computer Science from Institut Polytechnique de Paris (IP Paris) and a Bachelor’s degree in Computer Engineering from the Kwame Nkrumah University of Science and Technology. I am currently looking for internship opportunities in 2D/3D Vision,and Robotics. Feel free to contact me if I am a possible match for your team.

My research is centered around how we can build generalizable and robust algorithms for understanding and interacting with the visual world with minimal human supervision, making it possible to solve complex tasks in robotics, autonomous driving, and augmented and virtual reality. In pursuit of this objective, my research draws upon computer vision, representation learning, self-supervised learning, multi-modal learning, domain adaptation and natural language processing, seamlessly integrating these domains to develop computational models that deepen our understanding of the visual world and propel us closer to realizing this goal. Also, I aim to utilize the unexplored synergies between robot perception and foundation vision, language, and vision-language models, leveraging their inherent open-set capabilities to enhance the development of generalizable vision systems.

Interests

2D & 3D Computer Vision
Robotics (Perception, SLAM, navigation)

Education

Msc. Computer Science, 2023

Institut Polytechnique de Paris
BSc. Computer Engineering, 2020

Kwame Nkrumah University of Science and Technology

Recent News

Oct. 2023: I have received a MSc degree with highest honors and jury congratulations from Institut Polytechnique de Paris.
Sept. 2023: An extended abstract of our work on “How to train an ensemble of Faster R-CNN model to quantify uncertainity” at the 8th Junior Conference on Data Science and Engineering(JDSE) in Orsay, France 28 -29 September 2023.
Sept. 2023: I defended my Msc thesis at IP Paris. Thanks to Gilles Puy and David Hurych my advisors at valeo.ai.
Aug. 2023: I will be attending ICCV2023 in Paris, France as a student volunteer. I will be glad to connect with you.

Jan. 2023: I have accepted a research internship position for Spring and Summer 2023 at valeo.ai, a leading automonous vehicles R&D center in Paris. I will be working on domain adaptation for 3D semantic segmentation as the subject of my masters thesis.

Experience

Computer Vision Research Intern

Valeo AI

Mar 2023 – Aug 2023 Paris, France

Research Assistant

ENSTA PARIS, UI2S Research Laboratory

Sep 2021 – Feb 2023 Palaiseau,France

Computer Vision Research Intern

Tribun Health

May 2022 – Aug 2022 Paris,France

Software Engineer

Bewsys

Sep 2021 – Apr 2022 Accra,Ghana

Research and Teaching Assistant

Department of Computer Engineering, Kwame Nkrumah University of Science and Technology

Sep 2020 – Aug 2021 Kumasi,Ghana

Software Engineer

FillyCoder

Jul 2020 – Jul 2021 Accra,Ghana

Publications

A. Denis Mbey and F. Gianni, “How to effectively train an ensemble of Faster R-CNN object detectors to quantify uncertainty”, Pattern Analysis and Applications, 2022. - revised August 2023 arvix available here

Honors & Awards

Bourse Excellence scholarship receipent at IP Paris (2021-2023)
Campus France Ghana scholarship receipent at IP Paris (2021-2023)
Excellent student award recipient, College of Engineering, KNUST (2017,2018,2019)
Mastercard Foundation Scholarship award at KNUST (2016-2020) [Link]
Excellent graduating student, St. Charles Minor Seminary SHS (2015)

Projects

AITextDetector

This is a course project for Text Mining and NLP at Ecole Polytechnique aimed at building a classification model that can differentiate between AI-generated and human-written text. Pre-trained language models such as BERT, RoBERTa, and GPT-2 were utilized to train multiple models, with the best model achieving a 90% accuracy after extensive testing. The project has implications in critical domains such as journalism, social media, and customer service.

Uncertainity estimation in object detection models

This project aimed to quantify uncertainty in object detection models. To this end, I trained a Deep Ensemble of object detections models. Specifically, I proposed a new training protocol for training an ensemble of Faster R-CNN models using only one Region Proposal Network (RPN). In this training scheme, all you need is one RPN network and multiple Fast R-CNN prediction heads to create a robust ensemble network. This proposal produces results that are comparable to the traditional approach of fully training deep learning models end-to-end for ensembling.

Multiple Object Tracking

This project tackles multiple object tracking which is a challenging problem in Computer Vision. To this end,I trained a robust pedestrain object model using Faster R-CNN object detection algorithms to localise pedestrains in video frame sequences using MOT17DET dataset. I then leveraged tracking-by-detection algorithms such as SORT and Tracktor++ for tracking people in MOT17 dataset.

TelloSLAM

This project is an implementation of a SLAM package for the Tello drone using the ORBSLAM2 algorithm. The ROS package achieved near real-time SLAM results and enabled the Tello drone to build a map of an unknown environment and to simultaneously determine its location within the map.

H-index Prediction

H-index Prediction was a class based kaggle challenge with over 120 groups in the Machine and Deep Learning Course at Ecole Polytechnique. The objective was to find the best NLP model, based on the MSE (Mean Squared Error), to predict the h-index of scientific authors in the domain of computer science.

Image Filtering

This project exploit parallelism in a sequential image filtering program. This project aims to parallelize this program using parallel programming tools (OpenMP, MPI, and CUDA). This program reduced the computation time of applying Sobel and blur filters on gif images by 20%. The speed-up of image filtering programs is necessary as it is applicable in many sub-fields of image processing.

UAV path-planning and communication joint optimization

This project aimed to solve the non-convex optimization problem of finding an optimal path for a UAV to move from a base station to a target/inspection point while respecting the strict constraints of power, and data transmission rate. The problem was first transformed into a convex optimization problem and solved using classical optimization algorithms in CVXPY.

PlantDoc

PlantDoc is a deep learning solution for diagnosing and recommending treatment options for African food crops. The model currently identifies some diseases in apples, maize, blueberry, potato, tomato, and cherries. This is a work in progress, and I plan to include in this project a recommendation system to suggest treatment options for the identified diseases. Also, I would build a mobile application to make this model easily accessible to farmers so that they can use this application on any smart device.

HomDoc

HomDoc is a telehealth application that enables people to have access to medical care outside of their physical geographical region. This application incorporates call, chat, and video functionalities that allows patients to communicate with medical doctors over the internet. The application has helped to provide improved access to quality healthcare in Ghana and Nigeria.

Smart Irrigation System

I trained an LSTM deep learning model to predict rainfall patterns for Kumasi. I built an efficient and smart irrigation system to conserve water using predictions from the deep learning model. I also created a mobile app to provide real-time farm data to the farmer.

Community Engagement

Mentorship: I am always excited to mentor and support others to realize their full potential. I have mentored brilliant but needy students and guided them to apply for scholarships to pursue a university education. I have mentored five (5) students in my community, who are currently pursuing undergraduate programs at various universities in Ghana and abroad.

Community projects and Leadership: I founded a non-governmental organization called TechOne Foundation in Zebilla, my hometown in the Upper East Region of Ghana. Through this initiative, I teamed up with my friends in the community to provide ICT lessons to students. We have plans to extend the scope of the foundation to provide training programs in data science and software engineering to help the youth and school dropouts in Zebilla to acquire skills in these demanding areas to make themselves employable. Also, in France, I serve as the media secretary of the National Union of Ghanaians Students France Chapter.

Tutoring: During my time as a teaching assistant at the Department of Computer Engineering, I organized tutorial lessons in Computer Vision, Robotics, Linear Electronics, and Computer Graphics. In these sessions, I helped the students to have a further grasp of course concepts that were taught by the professors.

Volunteering: I have taken part in several community give-back initiatives during my time as a Mastercard Foundation Scholar. From teaching high school students to blood donation campaigns. I cherished every opportunity I had to give back to my community. In 2018, I spent two (2) months volunteering at Turing Trust Foundation in Kumasi, Ghana as a technical assitant. I installed useful educational software such as Microsoft Office Suite, Mavis Beacon Typing tutor and fixed glitches on recycled computers that were distributed to various basic and Junior High Schools in Ghanna to aid in the teaching and learning of ICT.

Please feel free to reach out, if you find this interesting and would like to collaborate for us to make an impact. I am always opened to learning from other people’s ideas. Together we can make the world better and put smiles on the faces of people.