Efficient Video Stitching Based on Fast Structure Deformation

Abstract

In computer vision, video stitching is a very challenging problem. In this paper, we proposed an efficient and effective wide-view video stitching method based on fast structure deformation that is capable of simultaneously achieving quality stitching and computational efficiency. For a group of synchronized frames, firstly, an effective double-seam selection scheme is designed to search two distinct but structurally corresponding seams in the two original images. The seam location of the previous frame is further considered to preserve the interframe consistency. Secondly, along the double seams, 1-D feature detection and matching is performed to capture the structural relationship between the two adjacent views. Thirdly, after feature matching, we propose an efficient algorithm to linearly propagate the deformation vectors to eliminate structure misalignment. At last, image intensity misalignment is corrected by rapid gradient fusion based on the successive over relaxation iteration (SORI) solver. A principled solution to the initialization of the SORI significantly reduced the number of iterations required. We have compared favorably our method with seven state-of-the-art image and video stitching algorithms as well as traditional ones. Experimental results show that our method outperforms the existing ones compared in terms of overall stitching quality and computational efficiency.

Real‐time mosaic of multiple fisheye surveillance videos based on geo‐registration and rectification

“…However, this assumption cannot guarantee the good alignment of the different fisheye videos after orthorectification, because there are always trees, vehicles, and other man-made objects on the ground, which will produce large object parallax and lead to obvious visual artefacts. On the other hand, as an elastic registration technology widely used in non-rigid medical image registration (Li et al, 2015), thin plate spline (TPS) can deal with the camera-model-independent spatial transformation of images by allowing local adaptation and constraint to certain continuity (Bookstein, 1989). To this end, this paper proposed an orthorectification constrained TPS technology to eliminate the serious distortion of fisheye videos and align them accurately, laying the foundation for seamless mosaic.…”

Stitching Videos from a Fisheye Lens Camera and a Wide-Angle Lens Camera for Telepresence Robots

“…These softwares usually compute a 2D transformation relating two cameras, and then bring all pairs of video images into alignment for the post-production of stitched videos. To improve the quality of the stitched video, Li et al [31] found double-seam to eliminate intensity misalignment, and similar work has been presented for designing a content-aware adaptive blending [26]. Some works were presented to obtain better alignment results.…”

A navigation paradigm driven classification for video-based rendering techniques

“…When adding a temporal dimension camera stabilization, new challenges have to be considered in the performed interpolation. Efficiency [44], color correction [45], wider baselines [46], ghosting artifacts [47], video stabilization [48] among others. These have been the main focus points in recent research, with each different algorithm and proposed technique being more suited to different type of content.…”