An Accurate Multi-Row Panorama Generation Using Multi-Point Joint Stitching

Most of the existing panorama generation tools require the input images to be captured along one direction, and yield a narrow strip panorama. To generate a large viewing field panorama, this paper proposes a multi-row panorama generation (MRPG) method. For a pan/tilt camera whose scanning path covers a wide range of horizontal and vertical views, the image frames in different views correspond to different coordinate benchmarks and different projections. And the image frame should not only be aligned with the continuous frames in timeline but also be aligned with other frames in spatial neighborhood even with long time intervals. For these problems, MRPG first designs an optimal scanning path to cover the large viewing field, and chooses the center frame as the reference frame to start to stitch. The stitching order of multi-frame is arranged in first-column and second-row to ensure a small alignment error. Moreover, MRPG proposes a multi-point joint stitching method to eliminate the seams and correct the distortions, which makes the current frame accurately integrated into the panoramic canvas from all directions. Experimental results show that MRPG can generate a more accurate panorama than other state-of-the-art image stitching methods, and give a better visual effect for a large viewing field panorama.

[1]  Fan Zhang,et al.  Parallax-Tolerant Image Stitching , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[2]  Yan Yan,et al.  A unified hypothesis generation framework for multi-structure model fitting , 2017, Neurocomputing.

[3]  Tat-Jun Chin,et al.  Simultaneously Fitting and Segmenting Multiple-Structure Data with Outliers , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Matthijs C. Dorst Distinctive Image Features from Scale-Invariant Keypoints , 2011 .

[5]  Andrew W. Fitzgibbon,et al.  Bundle Adjustment - A Modern Synthesis , 1999, Workshop on Vision Algorithms.

[6]  Jing Li,et al.  A Novel Visual-Vocabulary-Translator-Based Cross-Domain Image Matching , 2017, IEEE Access.

[7]  Chi-Keung Tang,et al.  Image Stitching Using Structure Deformation , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  James Davis,et al.  Mosaics of scenes with moving objects , 1998, Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231).

[9]  Yu-Sheng Chen,et al.  Natural Image Stitching with the Global Similarity Prior , 2016, ECCV.

[10]  Michael S. Brown,et al.  Constructing image panoramas using dual-homography warping , 2011, CVPR 2011.

[11]  Gérard G. Medioni,et al.  A graph-based global registration for 2D mosaics , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[12]  Matthew A. Brown,et al.  Automatic Panoramic Image Stitching using Invariant Features , 2007, International Journal of Computer Vision.

[13]  Tat-Jun Chin,et al.  Robust fitting of multiple structures: The statistical learning approach , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[14]  Sharath Pankanti,et al.  Adaptive as-natural-as-possible image stitching , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  S. Pravenaa A Methodical Review on Image Stitching and Video Stitching Techniques , 2016 .

[16]  Jean-Christophe Lapayre,et al.  An optimized algorithm of image stitching in the case of a multi-modal probe for monitoring the evolution of scars , 2013, Photonics West - Biomedical Optics.

[17]  Tomohiro Harada,et al.  Segmentation Mask Refinement Using Image Transformations , 2017, IEEE Access.

[18]  Yoichi Sato,et al.  Shape-Preserving Half-Projective Warps for Image Stitching , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[19]  Chi-Keung Tang,et al.  Eliminating structure and intensity misalignment in image stitching , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[20]  Michael S. Brown,et al.  As-Projective-As-Possible Image Stitching with Moving DLT , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.