Innovative Projects Realized

Indoor object detection for personal assistant for the blind

Humanware develops since 2003 a GPS-based dedicated device which helps a blind person to orient himself and guide him to his destination. In this project, we want to enhance the user’s experience by providing a better understanding of the environment by interacting with the device in a more natural manner. This project addresses two types of environment: the outdoor world where we want to improve door-to-door navigation by guiding the blind pedestrian to the door of the final destination and the indoor world where the device becomes a personal assistant.

Évaluation d’un acidifiant alimentaire chez le poulet de chair

La demande pour les produits de consommation sans médicament est en croissance constante. Cependant, certaines infections communes chez le poulet de chair affectent la santé et la production, exigeant l’utilisation d’alternatives aux médicaments. Parmi ces infections communes, la coccidiose aviaire engendre des pertes économiques considérables pour les producteurs. Pour l’instant, les alternatives efficaces aux anticoccidiens pour la production de poulet de chair sont manquantes. Même si des vaccins se sont développés et sont disponibles, leur efficacité n’est pas optimale et ils sont difficiles à exploiter en raison, notamment, du bas âge des poulets à l’abattage. Nous proposons d’étudier l’effet d’un acidifiant alimentaire chez le poulet de chair, en présence et en absence de coccidiose aviaire. Cette étude pourrait permettre d’offrir au secteur de la volaille du Canada une solution naturelle et efficace pour contrer la coccidiose.

Implementation of Environmental Social and Governance Initiatives of Canadian Mining Companies

The overall goal of the proposed research project is a creation of a database on corporate social responsibility (CSR) programs and broader Environmental, Social and Governance (ESG) initiatives sponsored by Canadian mining companies in the extractives sector, with a particular focus on programs directed to community development. The data aims to provide communities and other stakeholders with knowledge on community development and other programs implemented across the sector. The outcome will support the Canadian Executive Service Organization (CESO) in identifying areas where its expertise can be applied and assist in engaging with the industry on implementation of CSR and community engagement initiatives.

The Politics of Framing Youth and Doing Development in Nigeria

The research that I plan to conduct in this internship is part of my doctoral work. Through an ethnographic examination of the everyday realities of young men and women in West Africa’s largest marketplace, my PhD research will examine the relationship between livelihood opportunities, policy, and youth social action in Onitsha Eastern Nigeria. This research will ask: what kinds of external social, historical, and political forces are youth exposed to that shape their choices and life opportunities? Against the challenges that confront young people, what new forms of adaptation and resilience are being devised? How do young men and women work with and against established gendered forms of economic organization in African urban centers? How are they affected by neoliberalism and government policies? I will utilize this internship to advance the understanding of youthfulness and youth framing in Africa, arguing that effective youth intervention programming is contingent on a new conceptualization of what it means to a “youth” in Nigeria today. I argue that “being youth” in today’s Nigeria, and indeed in many parts of Africa, must be understood around certain key issues including information technology, migration, the dynamism of livelihood, among others.

Elaboration d’un système intégré de management des processus de production industrielle

Ce projet de recherche vise à tester des nouveaux modèles d’intégration des systèmes et de planification, de simulation, d’anticipation des opérations au sein de l’entreprise Sotrem-Maltech. Les retombées attendues sont de deux ordres : dans un premier temps l’entreprise bénéficiera d’une base d’information unique et partagée au sein des différents postes fonctionnels, dans un second temps, l’entreprise sera à jour des pratiques de planification opérationnelles pour mieux se positionner face à la concurrence mondiale.

Zipstall – On-line and Off-line Parking Availability Prediction

Searching for parking has many terrible impacts, such as wasted time, fuel, and emissions, overpaying for parking etc. To ease the pain of parking, the goal of this project is to develop a method of collecting information from multiple sources (crowd-sourced information from parkers, active paid session information from managers/parking enforcement, and availability information from enforcement patrols) and utilize various machine learning methods including K-Nearest Neighbors, Neural Network, Decision Tree and Time Series models to predict real-time on-street and off-street parking availability at the users’ estimated arrival time. Afterward, Zipstall can provide the users with personalized parking recommendation once the availability prediction of parking areas is accurate. The partner will benefit from participating in the program as advancements in the methodology of tracking and predicting real-time parking availability, which will enable the delivery of a vastly superior customer experience to the partner’s clients.

Differentially private models for detection of previously seen data

Jumio is constantly facing fraudulent attacks of repeated nature when a series of similar images with minor changes are submitted. To be able to respond effectively it is necessary to be able to learn previously seen fraudulent data. At the same time we have to deal with very sensitive private data and therefore it’s becoming a major concern for multiple reasons.

First, just comparing to all already seen data is not feasible technically due to optimization issues. And second most importantly it would cause legal issues due to privacy concerns and restrictions.

Predicting acoustic and pollutant emissions from combustion equipment using experiments and machine learning

Several engineering equipment ranging from those used for portable power generation, to medium scale gas turbine aircraft engines, and large scale land-based power generation units burn fuel to produce either electric energy or generate propulsive force. This energy conversion takes place inside a combustion chamber which emits noise and combustion pollutants. The objective of the present study is to first perform experiments and analyze data to understand the relations between noise and pollutant emissions. Then, artificial neural networks will be used to perform data mining, developing models that facilitate prediction of both noise and pollution emission from the combustors.

Object occlusion detection

Sometimes an object is partially hidden by a physical object like fingers, book etc. or by digitally pasted artifacts like blurringof an area. The intern will be working on detecting these types of digital and physical occlusions on an object.
At Jumio we encounter documents that are hidden or occluded by such masks. These documents need to be rejected or approved depending on the issuing country. The intern will help us develop a solution which can accurately detect any type of masks and helps us decide if we should approve or reject the masked document based on some decision rules. This project will directly contribute to our world-class product.

Portrait photo segmentation and generation with Deep Neural Network

In this project, the main purpose is to develop a tool for facial image anonymization by replacing the face in an image with a fake, generated face. To have realistic looking generated images, it is essential for the generated face to have a similar pose to the original one and to be seamlessly harmonic to the background. To achieve this goal, the first step is to locate all the faces of humans in images, and then extract key information about the pose (e.i.: eyebrows alignment) to reconstruct a similar face. Finally, a deep network will be trained to draw a fake face with extracted facial information on the location where faces are detected. This can be served as a data anonymization application protecting user’s privacy. In addition, it can be a tool for data augmentation which is a commonly used technique to boost deep neural networks’ performance.

Modelling of Passive Pilot, Pilot Seat and Inceptor for Aircraft-Pilot-Coupling (APC) Induced Oscillation Investigations

Aircraft Pilot Coupling (APC) may arise when airframe structural modes encroach into the frequency range of human senses, biodynamics and control, which is becoming more relevant given the increasingly flexible and optimized airframes of next-generation aircraft. The phenomenon is characterized by oscillations sensed by the aircraft crew and passengers that negatively impact the ride quality. Furthermore, it may impair the pilot ability to perform specific tasks and in some extreme cases might lead to fatalities. The project aims to develop and validate models representative of the relevant pilot biodynamics, aircraft inceptor and pilot seat that will be coupled to a supplied flexible aircraft model with flight control laws. The integrated model will be used to identify signatures of the APC phenomenon and understand the factors leading to its occurrence. The research project will provide Bombardier Aviation with the tool to predict potential APC in an aircraft development program to avoid costly design changes in its later stages, especially during flight testing. It will also allow for implementation and evaluation of solution strategies.

Applying state-of-art NLP models to molecular representation

Molecular generative methods are at the heart of our computational platform. We use cutting edge deep neural networks in order to generate unseen molecules based on existing conditions or molecular space. Since external projects typically require different architectures, we are continuously expanding our generative methods toolbox with new approaches. This project aims to augment our internal toolbox with large pretrained NLP models for molecular representation and generation. During this project, the intern will have to train and benchmark existing architecture with large datasets but also contribute/integrate new models such as the transformers provided by Hugging Face into our existing API.