Shape-Preserving Half-Projective Warps for Image Stitching

Abstract

This paper proposes a novel parametric warp which is a spatial combination of a projective transformation and a similarity transformation. Given the projective transformation relating two input images, based on an analysis of the projective transformation, our method smoothly extrapolates the projective transformation of the overlapping regions into the non-overlapping regions and the resultant warp gradually changes from projective to similarity across the image. The proposed warp has the strengths of both projective and similarity warps. It provides good alignment accuracy as projective warps while preserving the perspective of individual image as similarity warps. It can also be combined with more advanced local-warp-based alignment methods such as the as-projective-as-possible warp for better alignment accuracy. With the proposed warp, the field of view can be extended by stitching images with less projective distortion (stretched shapes and enlarged sizes).

Robust Multi-homography Method for Image Stitching under Large Viewpoint Changes

“…It allowed local deviations for better alignment accuracy. Shape preserving half projective (SPHP) warping method [9] spatially employed projective transformation for overlapping regions and similarity transformation for non-overlapping regions. Chang et al [10] combined the pixel-based method and feature-based method for image stitching, and followed by a shape preserving aggregation.…”

Spatially-Varying Image Warps for Scene Alignment

“…The goal of our method is related to multi-viewpoint panorama generation [8], [9], [10]. For reducing shape/area distortion while stitching multiple images, Chang et al [11] proposed the shape-preserving half-projective warp, a spatial combination of a projective transformation and a similarity transformation. It provides good alignment accuracy as projective warps while preserving the perspective of individual image as similarity warps.…”

Geometry-Based Global Alignment for GSMS Remote Sensing Images

“…The size of GSMS image is normalized to 10, 000 × 10, 000 pixels. Considering efficiency, both the GSHHG and GSMS images are divided into patches [35][36][37] whose size is S1 × S2 pixels. Furthermore, feature points are matched in each pair of patches.…”