Neural Augmentation Based Panoramic High Dynamic Range Stitching

TL;DR

This paper introduces a neural augmentation method for panoramic HDR stitching that combines physics-driven and data-driven approaches to produce artifact-free, high-quality panoramic images from multiple LDR inputs with different exposures.

Contribution

It proposes a novel neural augmentation algorithm that integrates physics-based OFOVs with data-driven refinement for improved panoramic HDR stitching.

Findings

Outperforms existing stitching algorithms in experiments.

Effectively handles saturated regions and large intensity changes.

Produces high-quality panoramic HDR images with fewer artifacts.

Abstract

Due to saturated regions of inputting low dynamic range (LDR) images and large intensity changes among the LDR images caused by different exposures, it is challenging to produce an information enriched panoramic LDR image without visual artifacts for a high dynamic range (HDR) scene through stitching multiple geometrically synchronized LDR images with different exposures and pairwise overlapping fields of views (OFOVs). Fortunately, the stitching of such images is innately a perfect scenario for the fusion of a physics-driven approach and a data-driven approach due to their OFOVs. Based on this new insight, a novel neural augmentation based panoramic HDR stitching algorithm is proposed in this paper. The physics-driven approach is built up using the OFOVs. Different exposed images of each view are initially generated by using the physics-driven approach, are then refined by a data-driven approach, and are finally used to produce panoramic LDR images with different exposures. All the panoramic LDR images with different exposures are combined together via a multi-scale exposure fusion algorithm to produce the final panoramic LDR image. Experimental results demonstrate the proposed algorithm outperforms existing panoramic stitching algorithms.