Work in Tech

Find your next role at Canada's fastest-growing tech companies

Super Resolution for Digital Elevation Models

National Research Council Canada

National Research Council Canada

Winnipeg, MB, Canada
Posted on Nov 20, 2024

Super Resolution for Digital Elevation Models - Research Affiliate Program

Reference number: RSN24J-098595-000473
Selection process number: 2024-RSN-EA-SPI-640561
Natural Resources Canada - Canada Center for Mapping and Earth Observation Ottawa
Winnipeg (Manitoba)
This is a student position of 15-20 hours per week. The hourly rate varies ($24.62 to $30.99) depending on the level of experience, according to the Treasury Board student rates of pay. Payments will be made in equal installments three times per year.

Closing date: 21 November 2024 - 23:59, Pacific Time

Who can apply: Persons residing in Canada and Canadian citizens residing abroad and international students.

To be considered for Research Affiliate Program (RAP) work opportunities, you must meet the following criteria:
• Be recognized as having full-time student status by the post-secondary academic institution at which you are currently enrolled (Proof will be required) or eligible to enroll in graduate program at University of Manitoba
• Be the minimum age to work in the province or territory where the job is located.

Apply online

Important messages

We are committed to providing an inclusive and barrier-free work environment, starting with the hiring process. If you need to be accommodated during any phase of the evaluation process, please use the Contact information below to request specialized accommodation. All information received in relation to accommodation will be kept confidential.

Assessment accommodation

Duties

The successful candidate will have the opportunity to conduct research in the field of geomatics, remote sensing and computer vision. The focus of this research concerns two elements related to super resolution technique for Digital Elevation Models (DEMs).
1) Generation of Accurate Training Data.
This research aims to develop methods for downsampling high-resolution Digital Elevation Models (DEMs) to generate training data for super-resolution models. Traditional downsampling techniques, such as bicubic downsampling, often fail to preserve critical topographic features like peaks and breaklines. Therefore, this study will investigate adaptive downsampling techniques to better maintain these important features in the training data.

2) Exploration of Transformer-Based Models for Super-Resolving DEMs.
This research focuses on leveraging transformer-based models to enhance the resolution of Digital Elevation Models (DEMs). The attention mechanism inherent in visual transformers (ViT) captures both local and global patterns, as well as relationships across different parts of the input. This capability is particularly advantageous for accurately super-resolving DEMs, as topographic features often extend over large areas. Furthermore, advanced visual transformers have demonstrated superior performance in super-resolution tasks compared to traditional CNN-based methods.
The successful student will have the opportunity to interact with CCMEO staff, researchers and technicians. The day-to-day work will take place at the University.

The successful candidate should have a good work ethic, a strong sense of responsibility, and the ability to work in team environments.

Preference will be given to a Canadian citizen and a permanent resident of Canada.

This position will start in May 2025.

Positions to be filled: 1

Information you must provide

Your résumé.

A covering letter "maximum with 2000 words"

Contact information for 2 references.

A list of the courses you have taken as well as any courses that you are taking now, or that you will be taking this academic year

In order to be considered, your application must clearly explain how you meet the following (essential qualifications)

Education: Currently enrolled or eligible to enroll in Master’s program in Computer Science, University of Manitoba.

Degree equivalency

Experience:
• Experience in analyzing and understanding geospatial data.
• Experience in accessing and preprocessing spatial data.
• Experience in remote sensing computer vision applications development in Python.
• Experience in conducting literature reviews and summarizing information.
• Experience in writing research reports and making scientific presentations.

The following will be applied / assessed at a later date (essential for the job)

English essential

Information on language requirements

Knowledge:
• Computer programming
• Knowledge of vector and raster data and 3D GIS
• Knowledge in digital elevation model
• Knowledge of computer vision techniques such as GANs, CNNs, ViT

Competencies:
• Interactive communication
• Initiative
• Adaptability
• Teamwork

Other information

The Public Service of Canada is committed to building a skilled and diverse workforce that reflects the Canadians we serve. We promote employment equity and encourage you to indicate if you belong to one of the designated groups when you apply.

Information on employment equity

For this selection process, it is our intention to communicate with candidates via email. Candidates must include a valid email address in their application. It is the candidate’s responsibility to ensure that this address is functional and that it accepts messages from unknown users (some email systems block these types of email).

A written examination may be administered.
An interview may be conducted.
Reference may be sought.

You must provide proof of your education credentials. Candidates with foreign credentials must provide proof of Canadian equivalency. Consult the Canadian Information Centre for International Credentials for further information at http://www.cicic.ca/.

Persons are entitled to participate in the appointment process in the official language of their choice.

You must indicate on your application if you require a technical aid for testing or an alternative method of assessment.

Candidates from outside the public service may be required to pay for travel and relocation costs associated with this selection process.

Successful completion of both a RAP work assignment and your educational program may lead to a temporary or permanent federal public service position for which you meet the merit criteria and conditions of employment.

Preference

Preference will be given to Canadian citizens and permanent residents, with the exception of a job located in Nunavut, where Nunavut Inuit will be appointed first.

We thank all those who apply. Only those selected for further consideration will be contacted.

Contact information

Christina Tarsky, HR Advisor
christina.tarsky@nrcan-rncan.gc.ca

Apply online