Computer Vision W21

The Waterloo Aerial Robotics Group (WARG) is an engineering student design team whose mission is to develop Unmanned Aerial Systems (UAS). WARG’s goal is to develop fully autonomous aircraft, which from takeoff to landing require absolutely no pilot control. The UAVs become autonomous to an extent where the aircraft can be controlled by a simple click on a map, or a touch on a smartphone. Complete paths can be created, altered, and  uploaded midflight in order to change the UAV's course of action. We aren’t quite there yet, and that’s why we need you!

 

WARG is ramping up for the 2021 Unmanned Systems Canada (USC) Student UAS competition, with the objective of simulating unmanned medical transport. You’ll be working on automated landing, geolocation, tracking and QR Identification with a dynamic team. Because we’re a software team, this role is completely remote and being in-person is not a requirement.

 

As a computer vision developer co-op, you’ll be working on every aspect of designing, developing and deploying a computer vision system from data ingestion to sending autopilot commands. Our system is still in its infancy, but we’ve got a solid roadmap and good foundations, we just need someone like you to help us tackle problems and build solutions. This will include data pipelines from our DeckLink video capture system, our YOLOv2 Neural Network, an object geolocation system, an API layer to send our results to the lovely autopilot system, and all the bells and whistles in between. This all takes some research, so be ready to read and learn. If you’re comfortable with Python, love solving new problems, have an understanding of machine learning concepts, and know how to Google effectively, we’d love to have you onboard at UWARG Computer Vision.

 

Required Skills

  • Familiar with Python or any other object-oriented, high-level language like Java, C#, JavaScript (so we’ll have an easy time converting you).

  • Familiarity with object-oriented concepts, such as classes, objects, encapsulation, interfaces and representing real-world systems with objects. 

  • You know how to turn a problem into a solution, and turn the solution into code.

  • Strong communication skills, capable of staying in touch, contributing to technical discussion and asking questions. 

  • A drive to learn, design teams are all about growth, and we want someone serious about growing their skills, technical or otherwise, at WARG.

 

Preferred Skills

  • Understanding of Machine Learning, specifically neural networks, especially convolutional neural networks. Can you talk me through how a neural network trains, without all the math? Bonus points if you can tell me how a ConvNet learns from images.

  • Understanding of computer video and image processing. Can you tell me how a computer stores images? 

  • Familiarity with OpenCV, Tensorflow/Keras and NumPy. 

  • Previous projects or experience in developing tools, utilities, or apps in Python. 

 

For more information, check out our website at: www.uwarg.com

***Please note that this role is unpaid***

Â