Out of Distribution Detection in Deep Generative Models

As generative models become increasingly prominent in machine learning, the need for accurately detecting out-of-distribution data has become crucial. The primary objective of this research is to develop an approach that can identify when the program encounters data that is vastly different from what it was trained on. In machine learning, programs may make errors […]

Read More
Transceivers based on quantum dash optical frequency comb sources for optical access networking

The project focuses on the design, modeling/simulation, and testing of photonic devices, and the co-design of electronicphotonic integrated circuits, to realize transceivers that can be commercialized for use in both wireline, e.g., passive optical networks (PONs), and wireless, e.g., 5G backhaul, access networks with an emphasis on meeting the reach requirements for remote and underserved […]

Read More
Adapting Not-for-Profit Arts Publishing to the Post-Digital Paradigm

PUBLIC (est. 1988) is a not-for-profit print periodical exploring the intersection of art, design, technology, and culture. This research project advances that history by looking at the future of print culture as today’s artists are increasingly making and exhibiting work in online spaces. How can PUBLIC promote engaging and accessible conversations around artists working exclusively […]

Read More
Object Tracking for High-Speed Pick-and-Place Robot

The demand for eCommerce and online orders has risen rapidly in recent years, this drives the need for highly efficient and automated item sortation systems. Kindred AI is a technology company with the objective to bring artificial intelligence and robotic technologies into the workforce of eCommerce, parcel and order fulfillment. As a part of the […]

Read More
Representation Learning with Time Series Data

The proposed research aims at learning better representations for multivariate time series (MTS) data, which can be applied to various important real-life applications such as weather, traffic, and electricity forecasting. Better forecasting accuracies for these tasks could help with efficient risk aversion and decision making, and save costs for decision makers. The proposed research will […]

Read More
Wearable Digital Technology for Continuous Monitoring of Vital Signs – part A

SFU will be collaborating with partners of a supercluster on the Continuous Connected Patient Care (CCPC) Platform, which is being developed to provide safe, responsive, and high-quality outpatient care at home. We will be working directly with our partner, Medtronic Canada. SFU’s will focus on the research, development, and testing of medical grade wearable sensor […]

Read More
Application of multi-agent systems and block chains in additive manufacturing network scheduling

Additive manufacturing technologies enable facilities at any location in the global supply chain to bid for customer jobs. In this setting, job scheduling is based on due-dates, proximity to customer location, opportunities for cheaper consolidated shipments with other customers, etc. Both multi-agent systems and block chains can support scheduling for decentralized manufacturing. In this project, […]

Read More
Application of Artificial Intelligence in Human Fall Detection

Falls are a significant public health concern, particularly among older adults, who are more vulnerable to injury and death resulting from falls. According to the World Health Organization (WHO), falls are the second leading cause of accidental or unintentional injury deaths worldwide. With the global population aging rapidly, there is an urgent need to develop […]

Read More
Paired-image enhancement for smartphone cameras

This project aims to develop enhancement techniques for high-resolution imaging on smartphone devices. The project will focus on a specific imaging modality where a pair of images are captured in succession: a high-resolution image (108 mega-pixel) and a corresponding low-resolution image (12 mega-pixel). Due to the small pixel size in the sensor used to capture […]

Read More
Bentley : Segmentation et analyse automatique de grands maillages 3D

De plus en plus de technologies existent pour capturer un modèle 3D d’une infrastructure existante : drones, robots, caméras spécialisées, etc. Toutefois, les données ainsi produites ne sont pas “intelligentes”, en ce sens qu’elles ne contiennent aucune information sur ce qu’elles représentent, pas plus que les pixels d’une image n’indiquent ce qu’ils représentent. Ce projet […]

Read More
Splittings of Binomial Edge Ideals

A graph is a collection of nodes or points, called vertices, along with a collection of objects called edges, which connect some of the vertices. Graphs can be used to model a number of real world applications. For example, the world wide web can be represented by a node for each website, and an edge […]

Read More