Computing correspondence based on regions and invariants without feature extraction and segmentation

The problem addressed in this paper is matching corresponding regions into images, even when the image intensity may be smoothly varying without distinctive edges. Corresponding small regions are assumed to be related by affine transformations. The matching is done by using a new class of low computational cost affine invariants. This approach also computes the affine transformation and is ideal for applications to 3D motion estimation and 3D surface reconstruction, image alignment, etc.. No feature extraction, segmentation or epipolar constraint is required. The very important advantage of our approach over area matching is that it handles large baselines, i.e., distance between camera positions where the differences in orientation and linear distortion of two areas being compared is large.