Post

DSC Conference Tutorials Day 2

In preparation for the DSC Conference we were offered 2 days of online webinar-style tutorials. On the second day i started off with a webinar about :

Object detection using OpenCV

Object Detection is The mix of classification (recognizing it is a cat) and localisation (seeing where on the image the cat is).

In this tutorial we explored the applications of object detection, going from turning on a light only when a person approaches (instead of whenever anything moves) to detecting road signalisation and traffic lights in self-driving cars.

We also explored many different methods used in object detection such as :

  • R-CNN (Region-based Convolutional Neural Networks)
  • Fast R-CNN
  • Faster R-CNN
  • YOLO (You Only Look Once)
  • SSD (Single Shot Detector)
  • RetinaNet
  • Mask R-CNN
  • Cascade R-CNN
  • CenterNet
  • EfficientDet

Then we dove deeper into OpenCV, a library of programming functions mainly aimed at real-time computer vision. We explored the different methods of object detection in OpenCV and how to use them :

  • Haar Cascade
  • Template Matching
  • Feature Matching
  • Contour Detection

And we finally tested out an application that detects faces, mouths and eyes in a picture.

Augmented Reality - ARKit framework showcase

This second tutorial was more of a showcase of the ARKit framework for iOS, where the presenter showed us his code line by line and gave a general explanation each time. The showcase was an application that detected your face and applied a mask on it, with the mask following your face movements.

This post is licensed under CC BY 4.0 by the author.