Matching a logo in two images

Now that we have a general idea of what FAST and BRIEF are, we can understand why the team behind ORB (composed of Ethan Rublee, Vincent Rabaud, Kurt Konolige, and Gary R. Bradski) chose these two algorithms as a foundation for ORB.

In their paper, the authors aim to achieve the following results:

  • The addition of a fast and accurate orientation component to FAST
  • The efficient computation of oriented BRIEF features
  • Analysis of variance and correlation of oriented BRIEF features
  • A learning method to decorrelate BRIEF features under rotational invariance, leading to better performance in nearest-neighbor applications

The main points are quite clear: ORB aims to optimize and speed up operations, including the very ...

Get Learning OpenCV 4 Computer Vision with Python 3 now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.