Implicit Neural Image Stitching

TL;DR

This paper introduces Implicit Neural Image Stitching (NIS), a novel method that enhances stitched image quality by estimating Fourier coefficients and blending features in latent space, overcoming limitations of previous approaches.

Contribution

The proposed NIS method extends arbitrary-scale super-resolution to improve image stitching quality by using Fourier coefficients and latent space blending, capturing high-frequency details.

Findings

Improves image quality over previous deep stitching methods

Effectively resolves low-definition and artifacts in stitched images

Achieves faster image enhancement with favorable results

Abstract

Existing frameworks for image stitching often provide visually reasonable stitchings. However, they suffer from blurry artifacts and disparities in illumination, depth level, etc. Although the recent learning-based stitchings relax such disparities, the required methods impose sacrifice of image qualities failing to capture high-frequency details for stitched images. To address the problem, we propose a novel approach, implicit Neural Image Stitching (NIS) that extends arbitrary-scale super-resolution. Our method estimates Fourier coefficients of images for quality-enhancing warps. Then, the suggested model blends color mismatches and misalignment in the latent space and decodes the features into RGB values of stitched images. Our experiments show that our approach achieves improvement in resolving the low-definition imaging of the previous deep image stitching with favorable accelerated image-enhancing methods. Our source code is available at https://github.com/minshu-kim/NIS.