Halder: Hierarchical Attention-Guided Learning with Detail-Refinement for Multi-Exposure Image Fusion

Liu, Jinyuan; Shang, Jingjie; Liu, Risheng; Fan, Xin

doi:10.1109/icme51207.2021.9428192

Cited by 6 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed method is compared with twelve state-of-theart methods, including four traditional methods, i.e., DSIFT [71], GBM [33], DEM [72] and MESPD [73], and eight deep learning-based methods, i.e., MEF-Net [74], IFCNN [75], MEF-GAN [47], U2Fusion [44], PMGI [76], HALDeR [68], CF-Net [48] and SD-Net [53]. Specifically, DSIFT used scale-invariant feature transform to extract local details and remove ghosting artifacts for fusion.…”

Section: Comparison Methodsmentioning

confidence: 99%

“…In the field of MEF, Structure Similarity Index Measure of Multi-exposure Image Fusion (MEF-SSIM) [43] and Peak Signal-to-Noise Ratio (PSNR) are introduced to conduct quantitative analysis, which are commonly used in abundant excellent works, e.g., [19], [47], [48] and [68]. At the same time, we also introduced Mutual Information (MI) [69] and Correlation Coefficient (CC) [70] as supplements.…”

Section: B Evaluation Metricsmentioning

confidence: 99%

See 1 more Smart Citation

Learning a Coordinated Network for Detail-Refinement Multiexposure Image Fusion

Liu

Zhou

et al. 2023

IEEE Trans. Circuits Syst. Video Technol.

Self Cite

View full text Add to dashboard Cite

Nowadays, deep learning has made rapid progress in the field of multi-exposure image fusion. However, it is still challenging to extract available features while retaining texture details and color. To address this difficult issue, in this paper, we propose a coordinated learning network for detail-refinement in an endto-end manner. Firstly, we obtain shallow feature maps from extreme over/under-exposed source images by a collaborative extraction module. Secondly, smooth attention weight maps are generated under the guidance of a self-attention module, which can draw a global connection to correlate patches in different locations. With the cooperation of the two aforementioned used modules, our proposed network can obtain a coarse fused image. Moreover, by assisting with an edge revision module, edge details of fused results are refined and noise is suppressed effectively. We conduct subjective qualitative and objective quantitative comparisons between the proposed method and twelve state-ofthe-art methods on two available public datasets, respectively. The results show that our fused images significantly outperform others in visual effects and evaluation metrics. In addition, we also perform ablation experiments to verify the function and effectiveness of each module in our proposed method. The source code can be achieved at https://github.com/lok-18/LCNDR.

show abstract