Search Results for: scale invariant

When matching points between scenes with large different viewpoints we need to account for varying image size and rotation. SIFT features are a powerful way to achieve this.

We use MATLAB and some Toolbox functions to find corresponding points between two images using SURF features.

We can describe the relationship between a 3D world point and a 2D image plane point, both expressed in homogeneous coordinates, using a linear transformation – a 3×4 matrix. Then we can extend this to account for an image plane which is a regular grid of discrete pixels.

We can also describe a blob by its contour or perimeter. Let’s look at how we determine the length of a blob’s perimeter using crack code and chain code. We can use the perimeter length to determine another scale and invariant shape parameter called circularity which indicates how compact, or circle-like, the blob is.

Image warping allows us to shrink (or expand) an image by any scale factor, as well as to translate and rotate it. Let’s look at how image warping works.

Image warping allows us to shrink (or expand) an image by any scale factor, as well as to translate and rotate it. Let’s look at how image warping works.

When it comes to describing a blob we can do more than just area, centroid position and bounding box. By looking at second order moments we can compute an ellipse that has the same moments of inertia as the blob, and we can use its aspect ratio and orientation to describe the shape and orientation […]

An image is a two dimensional projection of a three dimensional world. The big problem with this projection is that big distant objects appear the same size as small close objects. For people, and robots, it’s important to distinguish these different situations. Let’s look at how humans and robots can determine the scale of objects […]

Imagine trying to find a face in a crowd. If we know what the face looks like we could search for it at every possible location — this is the essence of template matching. To make it work we need to describe how similar each area we are checking is to the reference face image […]

The orientation of a body in 3D can also be described by two vectors, often called the approach and orientation vectors.