The Inverse Compositional Algorithm for Parametric Registration
Please cite the reference article if you publish results obtained with this online demo.

This demo interface performs the inverse compositional Lucas-Kanade method for computing a global transform between two images. It implements a coarse-to-fine strategy for estimating large displacements and uses robust functionals to deal with occlusions, noise and brightness changes.

Select Data

Click on an image pair to use it as the algorithm input.

identity
translation
zoom+rotation
zoom+rotat.+trans.
rotation+translation
rotation
zoom
homography 1
affinity
homography 2

Upload Data

Upload a pair of images to use as the algorithm input (and, optionally, a ground truth for comparison).




The ground truth must specify a two-dimensional vector field, for example a .flo file (as in the Middlebury database) or a TIFF file with two floating-point channels.