Local-Adaptive Image Alignment Based on Triangular Facet Approximation

Abstract

Accurate and efficient image alignment is the core problem in the research of panoramic stitching nowadays. This paper proposes a local-adaptive image alignment method based on triangular facet approximation, which directly manipulates the matching data in the camera coordinates, and therefore rises superior to the imaging model of cameras. A more robust planar transformation model is proposed and extended to be local-adaptive via combining it with two weighting strategies. By approximating the scene as a combination of adjacent triangular facets, the planar and spherical triangulation strategies are introduced to more efficiently align normal and fisheye images respectively. The efficiency of the proposed method is verified through the comparative experiments on several challenging cases both qualitatively and quantitatively.

Existing System

Content-Preserving Warps, Local-Adaptive Transformations

Proposed System

This paper proposed a local-adaptive image alignment method based on triangular facet approximation. First, the 8-parameter homographic model is converted to a more robust 3-parameter planar transformation model by manipulating the matching data in the camera coordinates which is irrelevant to the specific imaging model. Second, the planar transformation model is extended to be local-adaptive by applying the weighting strategies recommended by APAP and ANAP respectively. At last, according to the analysis on the limitations of the weighting methods, we approximate the scene as a combination of adjacent triangular facets and propose a triangulation-based local alignment method. The planar and spherical triangulation strategies are introduced for the alignment of perspective and fisheye images respectively. Triangular facet approximation method is more accurately aligned than those of global alignment and APAP. The proposed method rises superior to the imaging model of cameras, so that it applies to both perspective and non-perspective images. Without loss of alignment quality, the proposed triangular facet transformation drastically reduces both the number of local areas and involved matching data of each area

Architecture

BLOCK DIAGRAM