Learning Edge-Preserved Image Stitching from Large-Baseline Deep Homography

TL;DR

This paper introduces a novel deep learning framework for image stitching that effectively handles large-baseline transformations and preserves edges, resulting in improved accuracy and generalization over existing methods.

Contribution

The paper presents a large-baseline deep homography module and an edge-preserved deformation module, enabling robust image stitching across arbitrary views and input sizes.

Findings

Outperforms existing deep homography methods in large-baseline scenes.

Achieves superior stitching quality compared to existing learning-based methods.

Shows competitive performance with traditional state-of-the-art image stitching techniques.

Abstract

Image stitching is a classical and crucial technique in computer vision, which aims to generate the image with a wide field of view. The traditional methods heavily depend on the feature detection and require that scene features be dense and evenly distributed in the image, leading to varying ghosting effects and poor robustness. Learning methods usually suffer from fixed view and input size limitations, showing a lack of generalization ability on other real datasets. In this paper, we propose an image stitching learning framework, which consists of a large-baseline deep homography module and an edge-preserved deformation module. First, we propose a large-baseline deep homography module to estimate the accurate projective transformation between the reference image and the target image in different scales of features. After that, an edge-preserved deformation module is designed to learn the deformation rules of image stitching from edge to content, eliminating the ghosting effects as much as possible. In particular, the proposed learning framework can stitch images of arbitrary views and input sizes, thus contribute to a supervised deep image stitching method with excellent generalization capability in other real images. Experimental results demonstrate that our homography module significantly outperforms the existing deep homography methods in the large baseline scenes. In image stitching, our method is superior to the existing learning method and shows competitive performance with state-of-the-art traditional methods.