Projets novateurs réalisés

Explorez des milliers de projets réussis issus de la collaboration entre organisations et talents postsecondaires.

31 133 projets complétés

2940
AB
5159
C.-B.
837
MB
685
NL
882
SK
9292
ON
9695
QC
97
PE
601
NB
1161
NS

Projets par catégorie

Generation of a remote sensing methodology for the identification and variable chemical control of weeds with ground equipment in sugarcane

The proposed project intends to generate a method for identifying weeds and develop site-specific weed control prescriptions for sugarcane cultivations in Costa Rica using Remote Sensing techniques. Remotely sensed data will be captured using multispectral camera and LiDAR sensors attached to Remotely Piloted Aircraft Systems (RPAS). Images will be processed using machine learning algorithms to characterize the weed type and how it emerges during the early stages of sugarcane cultivation. Results will be evaluated using field samples. Once the variety of weeds and the emerging patterns identify, site-specific, variable-rate chemical solutions or prescriptions will be developed and fed into ground machines for spraying.

Voir la description complète du projet
Superviseur du corps professoral :

Muditha Heenkenda

Étudiant :

Partenaire :

National University of La Plata

Discipline :

Engineering

Secteur :

Environmental Science and Technology; Artificial Intelligence; Agriculture and Food

Université :

Lakehead University

Programme :

Globalink Research Award

Projet CCP : Cohabitation cyclistes-piétons sur les rues piétonnes : PARTIE 2

Pour répondre aux besoins des citoyens (mobilité, loisirs, activités sociales) en respectant les impératifs de distanciation physique, la Ville de Montréal et ses arrondissements ont implanté à l’été 2021 des rues piétonnes sur des artères commerciales, donc deux ont proposé des projets-pilotes de cohabitation des piétons avec les cyclistes (Avenue du Mont-Royal et rue Wellington). L’objectif du présent projet est de documenter cette cohabitation et d’en évaluer la sécurité (pour les piétons et les cyclistes) et l’acceptabilité sociale. Les analyses qualitatives et quantitatives qui seront effectuées dans le cadre de ces stages permettront de mieux comprendre la cohabitation entre les piétons et les cyclistes dans ces nouveaux espaces qui leur sont dédiés et d’ainsi proposer des améliorations à appliquer dans les prochaines années, sur ces deux rues ou sur d’autres qui voudraient adopter la cohabitation.

Voir la description complète du projet
Superviseur du corps professoral :

Marie-Soleil Cloutier;Nicolas Saunier;Francesco Ciari

Étudiant :

Partenaire :

Ville de Montréal

Discipline :

Sociology

Secteur :

Arts, entertainment and recreation; Public administration; Utilities

Université :

Polytechnique Montréal; Université du Québec : Institut national de la recherche scientifique

Programme :

Accelerate

Machine learning aided accelerated design and characterization of automotive composites

The proposed research project involves developing machine learning models to predict the mechanical properties of polymer composites. The interns will collect and preprocess data from various sources including open-source databases and conducting extensive experimental tests, build artificial neural network (ANN) models using advanced algorithms, and validate the accuracy of these models using test data. The expected benefit to the partner organization, Magna Closures, is the development of accurate and reliable models that can predict the behavior of polymer composites under different conditions, such as tensile tests and creep tests. These models can be used by the partner organization to optimize material selection and design, reduce testing costs, and improve the overall performance of polymer composites in various applications in automotive applications.

Voir la description complète du projet
Superviseur du corps professoral :

Reza Rizvi

Étudiant :

Partenaire :

Magna

Discipline :

Engineering

Secteur :

Manufacturing; Wholesale trade

Université :

York University

Programme :

Accelerate

Learning GPU Code structure using Transformers for Translation and Performance Optimization

This research project aims to improve the performance of GPUs, which are important for running machine learning algorithms. GPUs are a fundamental architecture in machine learning, and this project will use transformer-based models to learn the program structure of GPU kernels for various downstream tasks like performance projection and metrics such as GPU utilization. The research will also potentially allow bidirectional translation from assembly to source code, significantly improving code optimization and generation. The proposed research has the capacity to significantly improve code optimization and generation, leading to more sustainable and efficient practices and will also be beneficial to society for reducing the carbon footprint by reducing the number of optimizations runs.

Voir la description complète du projet
Superviseur du corps professoral :

Maryam Mehri Dehnavi

Étudiant :

Partenaire :

AMD Canada

Discipline :

Computer science

Secteur :

Manufacturing; Professional, scientific and technical services

Université :

University of Toronto

Programme :

Accelerate

Efficient Avatar Generation from Arbitrary Images

AR/VR may be the next frontier for online human communications and interactions. The ability to produce photorealistic avatars dramatically improves the feeling of immersion and connection in applications utilizing AR/VR. However, current methods of face capture are time-consuming and involve expensive cameras and sensors. In this project, we explore deep learning methods for generating face avatars using arbitrary images of a subject acquired on inexpensive consumer cameras, such as smartphone selfies, from various viewing angles and at different instances. Furthermore, the attributes of the generated avatars can be edited and animated. The project’s success will provide the partner organization with capabilities allowing it to build a new innovative product in the AR/VR space.

Voir la description complète du projet
Superviseur du corps professoral :

Karan Singh

Étudiant :

Partenaire :

AMD Canada

Discipline :

Computer science

Secteur :

Manufacturing; Professional, scientific and technical services

Université :

University of Toronto

Programme :

Accelerate

ML/AI LLVM Methods to Map Code to Core Architectures and Optimize

Modern compilers have increasingly large number of complex optimizations to meet the prevalent demand of using Machine Learning (ML) and Artificial Intelligence (AI) in gaming and other applications. Optimization passes are program and architecture depend. Therefore, selecting the best optimizations in the most optimal ordering is a difficult task. While leveraging ML methods in compiler optimization has become a prominent field of study, integration in production-level compiler for manycore architectures has yet to become standard. This project aims to find optimal methods for running sequential code onto the core architectures to find the most optimal performance and explore the trade-offs of performance versus ease of adoption by game developers of different solutions. The project would increase the hardware performance for ML/AI operations.

Voir la description complète du projet
Superviseur du corps professoral :

Nandita Vijaykumar

Étudiant :

Partenaire :

AMD Canada

Discipline :

Computer science

Secteur :

Manufacturing; Professional, scientific and technical services

Université :

University of Toronto

Programme :

Accelerate

Déploiement d’une politique de participation citoyenne : élaboration du modèle de la Ville de Magog

Quels critères doivent être considérés afin de déterminer les étapes à déployer pour optimiser les résultats des pratiques de consultation citoyenne? C’est la question centrale de ce projet de recherche développé en partenariat avec la ville de Magog.
Cette étude possède deux objectifs spécifiques. Premièrement, elle vise à accompagner cette municipalité située dans la région administrative de l’Estrie dans la mise en application de sa politique de participation citoyenne adoptée en septembre 2022. L’administration municipale s’interroge sur la façon d’adapter l’application de cette politique selon les types de propositions rencontrées. Ce premier objectif répond à des questions importantes liées à la mise en oeuvre de politiques de participation citoyenne : quand doit-on consulter? Pour quels projets? Dans chacun de ces cas, qui doit être consulté? Deuxièmement, cette recherche contribuera à la littérature entourant la mise en oeuvre des politiques publiques et la communication stratégique au palier municipal en particulier. Une meilleure compréhension des enjeux d’acceptabilité sociale au sein des collectivités découlera aussi de la réalisation de cette étude.

Voir la description complète du projet
Superviseur du corps professoral :

Joanie Bouchard;Emmanuel Choquette

Étudiant :

Partenaire :

Ville de Magog

Discipline :

Sociology

Secteur :

Public administration

Université :

Université de Sherbrooke

Programme :

Accelerate

Recherche et développement d’un modèles de prédiction de niveau de nappe phréatique pour une plateforme d’aide à la décision de type SaaS pour une application d’optimisation environnementale

THIS IS A GENERIC TEXT PUT IN PLACE AS THERE WAS NO PROJECT OVERVIEW

Voir la description complète du projet
Superviseur du corps professoral :

Ioannis Mitliagkas

Étudiant :

Partenaire :

NORDIKeau Inc

Discipline :

Computer science

Secteur :

Transportation and warehousing

Université :

Université de Montréal

Programme :

Accelerate

AI-Based Content Adaptive Video Compression

The rapid evolution of video resolution has significantly increased the video bitrate requirement, making data transfer a challenging task for data-intensive applications like video conferencing, cloud gaming and game streaming. With the rise of machine learning, studies have shown the potential of embedding conventional video compression algorithms with AI-based methods to enhance their performance. This research project aims to explore the potential features from the input video that can be leveraged by machine learning algorithms to predict the optimal parameters used in the video compression process, with the goal of maximizing the quality of the decompressed video under fixed bitrate constrain.

Voir la description complète du projet
Superviseur du corps professoral :

Qiang Sun

Étudiant :

Partenaire :

AMD Canada

Discipline :

Computer science

Secteur :

Manufacturing; Professional, scientific and technical services

Université :

University of Toronto

Programme :

Accelerate

Multi-task Reinforcement Learning for Video Games

An important component of modern video games is the non-player character (NPC), moving entities in-game that are not controlled by a human, which may cooperate with, oppose, or otherwise interact with the player. For an NPC to interact with the game word it must often perform complex tasks that are difficult to program explicitly. Research has explored using artificial intelligence, particularly reinforcement learning, to train NPCs to achieve the desired behavior, but prior work has often focused on training NPCs only within one game. Our goal is to investigate using multi-task reinforcement learning to train NPCs that are more robust and can easily transfer from one gaming task to another without needing to be retrained. Success could, in the long term, lead to downstream development of artificial intelligence that can transfer reliably between different use cases.

Voir la description complète du projet
Superviseur du corps professoral :

Florian Shkurti

Étudiant :

Partenaire :

AMD Canada

Discipline :

Computer science

Secteur :

Manufacturing; Professional, scientific and technical services

Université :

University of Toronto

Programme :

Accelerate

Multimodal Game Event Detection via Machine Learning

The partner company (AMD) is a major innovator in the field of computer graphics and visualization, they manufacture Graphical Processing Unit (GPU) which are used by many gamers around the world. While playing video games, gamers tend to perform out-of-band actions such as saving the last few minutes of gameplay after a challenging fight in a first-person shooter game. Gamers usually search for walkthroughs or FAQs when failing to complete a difficult level or scene in a video game. This project aims to use Computer Vision and Machine learning to detect such events in near real-time during a gameplay. Detection of such events can then trigger actions via AMD’s software to improve the gameplay experience of gamers who are using AMD hardware, thus providing benefit to their customers.

Voir la description complète du projet
Superviseur du corps professoral :

Chris Mcintosh

Étudiant :

Partenaire :

AMD Canada

Discipline :

Computer science

Secteur :

Manufacturing; Professional, scientific and technical services

Université :

University of Toronto

Programme :

Accelerate

Understanding the concept of ehealth literacy

eHealth literacy is an evolving concept. It influences the development of the content of digital interventions, how individuals interact with the information they receive, and how programmers and information scientists adapt their designs. However, the challenge is that it has no universal definition nor measuring tool. This challenge makes it difficult to compare and communicate the outcomes and results of studies because researchers often choose the concept and approach that fits their research. Findings show that current concepts seem inadequate as they were developed before the recent emergence of Web 3.0 and 4.0. As such, there is a need to create a standardised definition of eHealth literacy, identify or develop a robust measurement tool, and evaluate the association between eHealth literacy and adherence to behavioural changes in response to eHealth interventions. For the 12 weeks internship, we hope to identify relevant definitions from literature and synthesise them to develop a standardised definition.

Voir la description complète du projet
Superviseur du corps professoral :

Simon Bacon

Étudiant :

Partenaire :

University of Birmingham

Discipline :

Sociology

Secteur :

Health and Related Sciences & Technology; Information and Communications Technology

Université :

Concordia University

Programme :

Globalink Research Award