Unsupervised Multi-Exposure Image Fusion Breaking Exposure Limits via Contrastive Learning

Authors: Han Xu, Liang Haochen, Jiayi Ma

AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Qualitative, quantitative, and ablation experiments validate the superiority and generalization of MEF-CL. Our code is publicly available at https://github.com/hanna-xu/MEF-CL.
Researcher Affiliation Academia Electronic Information School, Wuhan University, Wuhan 430072, China
Pseudocode No The paper describes network architectures and processes but does not include a clearly labeled pseudocode block or algorithm.
Open Source Code Yes Our code is publicly available at https://github.com/hanna-xu/MEF-CL.
Open Datasets Yes We conduct experiments on the SICE dataset (Cai, Gu, and Zhang 2018)1 and perform the verification on different scenes, including indoor and outdoor scenes.1https://github.com/csjcai/SICE
Dataset Splits Yes We randomly selected 479 image sequences as the training set. The remaining 80 image sequences are as the test set.
Hardware Specification Yes The experiments are performed on an NVIDIA Geforce GTX Titan V GPU. Traditional methods are tested on a laptop with 3.2 GHz AMD Ryzen 7 5800H CPU.
Software Dependencies No The paper mentions 'Tensor Flow' but does not specify its version number or other software dependencies with versions.
Experiment Setup Yes The hyper-parameters are set as: λ1 = 10, λ2 = 20, τ = 0.01. The batch size is set to 20, the training epoch is 2, and the learning rate is 0.0001. We use the RMSProp optimizer for optimization.