Attention-Guided Progressive Neural Texture Fusion for High Dynamic Range Image Restoration

TL;DR

This paper introduces an advanced HDR image restoration model that employs attention-guided neural texture fusion to effectively address saturation, motion, and artifacts, outperforming existing methods.

Contribution

The proposed APNT-Fusion model uniquely integrates multi-scale neural feature transfer, progressive texture blending, and novel attention mechanisms for improved HDR image restoration.

Findings

Outperforms state-of-the-art HDR fusion methods in qualitative and quantitative metrics.

Effectively suppresses ghosting, noise, and blur artifacts.

Demonstrates robust performance across various challenging HDR scenarios.

Abstract

High Dynamic Range (HDR) imaging via multi-exposure fusion is an important task for most modern imaging platforms. In spite of recent developments in both hardware and algorithm innovations, challenges remain over content association ambiguities caused by saturation, motion, and various artifacts introduced during multi-exposure fusion such as ghosting, noise, and blur. In this work, we propose an Attention-guided Progressive Neural Texture Fusion (APNT-Fusion) HDR restoration model which aims to address these issues within one framework. An efficient two-stream structure is proposed which separately focuses on texture feature transfer over saturated regions and multi-exposure tonal and texture feature fusion. A neural feature transfer mechanism is proposed which establishes spatial correspondence between different exposures based on multi-scale VGG features in the masked saturated HDR domain for discriminative contextual clues over the ambiguous image areas. A progressive texture blending module is designed to blend the encoded two-stream features in a multi-scale and progressive manner. In addition, we introduce several novel attention mechanisms, i.e., the motion attention module detects and suppresses the content discrepancies among the reference images; the saturation attention module facilitates differentiating the misalignment caused by saturation from those caused by motion; and the scale attention module ensures texture blending consistency between different coder/decoder scales. We carry out comprehensive qualitative and quantitative evaluations and ablation studies, which validate that these novel modules work coherently under the same framework and outperform state-of-the-art methods.