Abstract
In this paper, we propose a unified framework to generate a pleasant and high-quality street-view panorama by stitching multiple panoramic images captured from the cameras mounted on the mobile platform. Our proposed framework is comprised of four major steps: image warping, color correction, optimal seam line detection and image blending. Since the input images are captured without a precisely common projection center from the scenes with the depth differences with respect to the cameras to different extents, such images cannot be precisely aligned in geometry. Therefore, an efficient image warping method based on the dense optical flow field is proposed to greatly suppress the influence of large geometric misalignment at first. Then, to lessen the influence of photometric inconsistencies caused by the illumination variations and different exposure settings, we propose an efficient color correction algorithm via matching extreme points of histograms to greatly decrease color differences between warped images. After that, the optimal seam lines between adjacent input images are detected via the graph cut energy minimization framework. At last, the Laplacian pyramid blending algorithm is applied to further eliminate the stitching artifacts along the optimal seam lines. Experimental results on a large set of challenging street-view panoramic images captured form the real world illustrate that the proposed system is capable of creating high-quality panoramas.
Highlights
Nowadays, with the development of street-view panoramas, which provide 360◦ panoramic views along streets in the real world, the demand for high-quality panoramic images gradually becomes greater
In our previous work presented in [19], we proposed an efficient optimal seam line detection algorithm for mosaicking aerial and panoramic images based on the graph cut energy minimization framework
When the geometric misalignments are very large, the stitching artifacts perhaps cannot be completely avoided even though the optimal seam lines are detected, especially for street-view panoramic images among which there always exist geometric misalignments to different extents due to those images being captured from scenes with large depth differences by a panoramic camera comprised of multiple wide-angle or fish-eye cameras without a precisely common projection center, which means that the geometric misalignments are different at different positions
Summary
With the development of street-view panoramas, which provide 360◦ panoramic views along streets in the real world, the demand for high-quality panoramic images gradually becomes greater. When the geometric misalignments are very large, the stitching artifacts perhaps cannot be completely avoided even though the optimal seam lines are detected, especially for street-view panoramic images among which there always exist geometric misalignments to different extents due to those images being captured from scenes with large depth differences by a panoramic camera comprised of multiple wide-angle or fish-eye cameras without a precisely common projection center, which means that the geometric misalignments are different at different positions. We adopt an efficient seam line detection approach based on the graph cut energy minimization framework to find the optimal seam lines between two overlapped images followed by applying the image blending to eliminate the color transitions along the seam lines.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
