As-Projective-As-Possible Image Stitching with Moving DLT

Abstract

We investigate projective estimation under model inadequacies, i.e., when the underpinning assumptions of the projective model are not fully satisfied by the data. We focus on the task of image stitching which is customarily solved by estimating a projective warp -a model that is justified when the scene is planar or when the views differ purely by rotation. Such conditions are easily violated in practice, and this yields stitching results with ghosting artefacts that necessitate the usage of deghosting algorithms. To this end we propose as-projective-as-possible warps, i.e., warps that aim to be globally projective, yet allow local non-projective deviations to account for violations to the assumed imaging conditions. Based on a novel estimation technique called Moving Direct Linear Transformation (Moving DLT), our method seamlessly bridges image regions that are inconsistent with the projective model. The result is highly accurate image stitching, with significantly reduced ghosting effects, thus lowering the dependency on post hoc deghosting.

Accurate and robust feature-based homography estimation using HALF-SIFT and feature localization error weighting

“…This shows that the effects of localization error on standard feature-based homography estimation can be avoided effectively and the accuracy and robustness of homography estimation can be improved significantly by using the proposed method. Therefore, the performance of many applications such as image stitching [4], image mosaicking [29] and augmented reality [30] adopting homography estimation technique can be improved by using the proposed method.…”

“…Homography plays an important role in many applications: camera calibration [1], 3D reconstruction, image registration [2,3], image stitching [4] and feature tracking.…”

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.…”

Diverse Scene Stitching from a Large-Scale Aerial Video Dataset

“…Image stitching has been well studied over the past decade [1][2][3][4][5][6][7][8][9][10]. We refer the readers to the comprehensive survey [11] for a background.…”