Till this point, we have used global features like color and shape to detect an object. These features are easy to compute, are quick, and require a small amount of memory, but they can only be used when some information regarding the object is already available. If that is not the case then local features are used, which require more computation and memory, but they are more accurate. In this section, various algorithms that find local features are explained. They are also called key point detectors. Key-points are the points that characterize the image and can be used to define an object accurately.
Key-point detectors and descriptors
Features from Accelerated Segment Test (FAST) feature detector
The FAST algorithm is used...
