C H A P T E R  8

Point Clouds, Part 2

This chapter builds on what you learned about point clouds and the Kinect to explore more advanced 3-D concepts. We will extend our methods to handle more than one image at the time. There are many things we could do with a point cloud, and in this chapter, we will explore three areas:

  • Registration
  • Simultaneous localization and mapping (SLAM)
  • Surface reconstruction

Constructing a point cloud from the depth and color images of the Kinect is only the first step. For most applications, we must still do some processing to be able to use the point cloud. So far, the point clouds we’ve constructed from each frame are independent from each other. It would be very useful if we could relate the different point ...

Get Hacking the Kinect now with O’Reilly online learning.

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