Wide Baseline Stereo Matching based on Local, Affinely Invariant Regions

TLDR

Invariant regions are image patches that automatically deform with viewpoint changes to cover identical physical parts, described by invariant features that enable easy matching across views and illumination, and prior work presented such regions based on corners and edges. The authors propose a purely intensity‑based method for extracting affinely invariant regions and demonstrate its use for wide‑baseline stereo matching, aiming to build an opportunistic system that exploits multiple invariant region types as needed. The system incorporates two semi‑local constraints—geometric and photometric—on region correspondence combinations to test consistency and reject false matches. The method yields more correspondences and a system that can handle a wider range of images, extending its applicability beyond prior image database retrieval.

Abstract

‘Invariant regions’ are image patches that automatically deform with changing viewpoint as to keep on covering identical physical parts of a scene. Such regions are then described by a set of invariant features, which makes it relatively easy to match them between views and under changing illumination. In previous work, we have presented invariant regions that are based on a combination of corners and edges. The application discussed then was image database retrieval. Here, an alternative method for extracting (affinely) invariant regions is given, that does not depend on the presence of edges or corners in the image but is purely intensity-based. Also, we demonstrate the use of such regions for another application, which is wide baseline stereo matching. As a matter of fact, the goal is to build an opportunistic system that exploits several types of invariant regions as it sees fit. This yields more correspondences and a system that can deal with a wider range of images. To increase the robustness of the system even further, two semi-local constraints on combinations of region correspondences are derived (one geometric, the other photometric). They allow to test the consistency of correspondences and hence to reject falsely matched regions.