Chapter 4. Exploring Structure from Motion Using OpenCV

In this chapter we will discuss the notion of Structure from Motion (SfM), or better put as extracting geometric structures from images taken through a camera's motion, using functions within OpenCV's API to help us. First, let us constrain the otherwise lengthy footpath of our approach to using a single camera, usually called a monocular approach, and a discrete and sparse set of frames rather than a continuous video stream. These two constrains will greatly simplify the system we will sketch in the coming pages, and help us understand the fundamentals of any SfM method. To implement our method we will follow in the footsteps of Hartley and Zisserman (hereafter referred to as H and Z), as documented in chapters 9 through 12 of their seminal book Multiple View Geometry in Computer Vision.

In this chapter we cover the following:

• Structure from Motion concepts

• Estimating the camera motion from a pair of images

• Reconstructing the scene

• Reconstruction...