HDRTransDC: High Dynamic Range Image Reconstruction with Transformer Deformation Convolution

TL;DR

HDRTransDC leverages transformer-based deformable convolution and dynamic fusion to produce high-quality HDR images from multi-exposure LDR inputs, effectively reducing ghosting and fusion artifacts.

Contribution

This paper introduces a novel HDR reconstruction network combining transformer deformable convolution and adaptive fusion, addressing misalignment and distortion issues in HDR imaging.

Findings

Achieves state-of-the-art HDR reconstruction quality.

Effectively reduces ghosting artifacts and fusion distortions.

Demonstrates superior performance in extensive experiments.

Abstract

High Dynamic Range (HDR) imaging aims to generate an artifact-free HDR image with realistic details by fusing multi-exposure Low Dynamic Range (LDR) images. Caused by large motion and severe under-/over-exposure among input LDR images, HDR imaging suffers from ghosting artifacts and fusion distortions. To address these critical issues, we propose an HDR Transformer Deformation Convolution (HDRTransDC) network to generate high-quality HDR images, which consists of the Transformer Deformable Convolution Alignment Module (TDCAM) and the Dynamic Weight Fusion Block (DWFB). To solve the ghosting artifacts, the proposed TDCAM extracts long-distance content similar to the reference feature in the entire non-reference features, which can accurately remove misalignment and fill the content occluded by moving objects. For the purpose of eliminating fusion distortions, we propose DWFB to spatially adaptively select useful information across frames to effectively fuse multi-exposed features. Extensive experiments show that our method quantitatively and qualitatively achieves state-of-the-art performance.