Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Applied Research and Development Grants - Level 1

Sheridan Institute of Technology and Advanced Learning

$25,000.00

Synthetic Dataset Creation

EF 2018-19

Résumé
Spxtrm AI is a Canadian based software company committed to improving elder care and safety by advancing leading-edge software capable of autonomously monitoring seniors and other high-risk patient populations. As the median age of Canadians continues to climb, intelligent monitoring systems are becoming increasingly attractive as cost-effective means of ensuring the safety and independence of elders and others. A critical challenge is the lack of robust datasets of recognizable human actions (i.e. video images of falls, or of actions that relate to elder abuse) that can be used to train a computer vision system to accurately recognize these actions. Spxtrm seeks to collaborate with a research team at Sheridan College's Screen Industries Research and Training Centre (SIRT) to develop a cost-effective process for creating ""synthetic datasets"". Researchers will determine if techniques used in motion capture for film production can be adapted to address dataset creation. Using virtual cameras, researchers will multiply the output of limited recorded actions, creating multiple viewing angles for a given action that can be used to ""train"" Spxtrm's computer vision system. SIRT and Spxtrm aim to develop best practices for capture, processing and rendering of motion capture data to minimize creation time and maximize output. Devising a cost-effective way to create difficult-to-obtain datasets is a huge business opportunity for Spxtrm as the solution can be tailored to address individual institution's key monitoring issues, while achieving a much higher classification and recognition rate than is possible with existing solutions. The ability to economically create synthetic datasets has potential for immediate payback as it would dramatically reduce time-to-market of solutions and machine learning training time/setup costs, all while improving recognition accuracy. This promises significant quality of life improvements for Canadian seniors, while helping combat rising per-capita healthcare costs associated with care for the aging. Global market potential for Spxtrm's proposed innovation is significant and will help position Canada as a world leader in providing better elder care through autonomous computer vision monitoring.*
Programme
Applied Research and Development Grants - Level 1
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
533397-2018
Name nom
England, Andrea
Department département
Head Office
Organization id
22156
Institution établissement
Sheridan Institute of Technology and Advanced Learning
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2018
Competition year année de concours
2018
Award amount
25000
Program id
CARD1
Program name en
Applied Research and Development Grants - Level 1
Program name fr
Subventions de recherche et développement appliquée - niveau 1
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1561
Committee name en
Applied Research and Development Grants
Committee name fr
Subventions de recherche et développement appliquée
Area of application code
801
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer software
Area of application fr
Logiciel
Research subject code
2800
Research subject group en
Artificial intelligence
Research subject group fr
Intelligence artificielle
Research subject en
Artificial intelligence (computer vision, use2603)
Research subject fr
Intelligence artificielle (vision artificielle, utiliser 2603)
Partie
1
Nb partie
1
Application title
Synthetic Dataset Creation
Keyword
NSERC
Application summary
Spxtrm AI is a Canadian based software company committed to improving elder care and safety by advancing leading-edge software capable of autonomously monitoring seniors and other high-risk patient populations. As the median age of Canadians continues to climb, intelligent monitoring systems are becoming increasingly attractive as cost-effective means of ensuring the safety and independence of elders and others. A critical challenge is the lack of robust datasets of recognizable human actions (i.e. video images of falls, or of actions that relate to elder abuse) that can be used to train a computer vision system to accurately recognize these actions. Spxtrm seeks to collaborate with a research team at Sheridan College's Screen Industries Research and Training Centre (SIRT) to develop a cost-effective process for creating ""synthetic datasets"". Researchers will determine if techniques used in motion capture for film production can be adapted to address dataset creation. Using virtual cameras, researchers will multiply the output of limited recorded actions, creating multiple viewing angles for a given action that can be used to ""train"" Spxtrm's computer vision system. SIRT and Spxtrm aim to develop best practices for capture, processing and rendering of motion capture data to minimize creation time and maximize output. Devising a cost-effective way to create difficult-to-obtain datasets is a huge business opportunity for Spxtrm as the solution can be tailored to address individual institution's key monitoring issues, while achieving a much higher classification and recognition rate than is possible with existing solutions. The ability to economically create synthetic datasets has potential for immediate payback as it would dramatically reduce time-to-market of solutions and machine learning training time/setup costs, all while improving recognition accuracy. This promises significant quality of life improvements for Canadian seniors, while helping combat rising per-capita healthcare costs associated with care for the aging. Global market potential for Spxtrm's proposed innovation is significant and will help position Canada as a world leader in providing better elder care through autonomous computer vision monitoring.*