Deep High Dynamic Range Imaging with Large Foreground Motions

TL;DR

This paper introduces a novel deep learning framework for HDR imaging of dynamic scenes with large foreground motions that does not rely on optical flow, effectively handling occlusion, saturation, and under-exposure.

Contribution

It presents the first non-flow-based deep HDR imaging method capable of hallucinating details in challenging conditions, outperforming flow-based approaches.

Findings

Produces high-quality HDR images with fewer artifacts.

Robust across various input types, including uncalibrated images.

Outperforms existing methods in qualitative and quantitative evaluations.

Abstract

This paper proposes the first non-flow-based deep framework for high dynamic range (HDR) imaging of dynamic scenes with large-scale foreground motions. In state-of-the-art deep HDR imaging, input images are first aligned using optical flows before merging, which are still error-prone due to occlusion and large motions. In stark contrast to flow-based methods, we formulate HDR imaging as an image translation problem without optical flows. Moreover, our simple translation network can automatically hallucinate plausible HDR details in the presence of total occlusion, saturation and under-exposure, which are otherwise almost impossible to recover by conventional optimization approaches. Our framework can also be extended for different reference images. We performed extensive qualitative and quantitative comparisons to show that our approach produces excellent results where color artifacts and geometric distortions are significantly reduced compared to existing state-of-the-art methods, and is robust across various inputs, including images without radiometric calibration.