Learning Residual Elastic Warps for Image Stitching under Dirichlet Boundary Condition
This work addresses a specific issue in image stitching for computer vision applications, offering an incremental improvement over existing methods.
The paper tackles the problem of holes and discontinuity in learning-based elastic warps for image stitching under large parallax errors by proposing REwarp, which uses Dirichlet boundary conditions and residual learning to achieve favorable alignments with competitive computational costs.
Trendy suggestions for learning-based elastic warps enable the deep image stitchings to align images exposed to large parallax errors. Despite the remarkable alignments, the methods struggle with occasional holes or discontinuity between overlapping and non-overlapping regions of a target image as the applied training strategy mostly focuses on overlap region alignment. As a result, they require additional modules such as seam finder and image inpainting for hiding discontinuity and filling holes, respectively. In this work, we suggest Recurrent Elastic Warps (REwarp) that address the problem with Dirichlet boundary condition and boost performances by residual learning for recurrent misalign correction. Specifically, REwarp predicts a homography and a Thin-plate Spline (TPS) under the boundary constraint for discontinuity and hole-free image stitching. Our experiments show the favorable aligns and the competitive computational costs of REwarp compared to the existing stitching methods. Our source code is available at https://github.com/minshu-kim/REwarp.