8.Motion Estimation
Motion estimation is an important topic in image processing, especially in computer vision. Being able to estimate the motion between two images of a sequence opens a field of applications in object tracking, image registration, 3D reconstruction, or even video compression.
Motion analysis can be categorized into three different classes of methods:
3D-3D approaches, which involve computing 3D motion from a set of 3D correspondences. Generally, 3D data are difficult to obtain;
2D-3D approaches, which estimate the 3D motion from correspondences between a 3D model and 2D projections;
2D-2D approaches, which do not use a 3D model and just assume correspondences between 2D projections, from which the 3D motion can be reconstructed. ...
Get Digital Image Processing with C++ now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
