Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].

EMEF: Ensemble Multi-Exposure Image Fusion

Authors: Renshuai Liu, Chengyang Li, Haitao Cao, Yinglin Zheng, Ming Zeng, Xuan Cheng

AAAI 2023 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In the experiment, we construct EMEF from four state-of-the-art MEF methods and then make comparisons with the individuals and several other competitive methods on the latest released MEF benchmark dataset. The promising experimental results demonstrate that our ensemble framework can get the best of all worlds .
Researcher Affiliation Academia Renshuai Liu, Chengyang Li, Haitao Cao, Yinglin Zheng, Ming Zeng, Xuan Cheng* School of Informatics, Xiamen University, Xiamen 361005, China EMAIL, EMAIL
Pseudocode Yes Algorithm 1: Search for the optimal style code c Rn
Open Source Code Yes The code is available at https://github.com/medalwill/EMEF.
Open Datasets Yes We train EMEF with the SICE (Cai, Gu, and Zhang 2018) dataset and evaluate it in MEFB (Zhang 2021).
Dataset Splits No The paper specifies training data and evaluation datasets but does not explicitly provide details for a validation split, nor exact percentages or counts for distinct training, validation, and test partitions for its model training.
Hardware Specification Yes All experiments are conducted with two Ge Force RTX 3090 GPUs.
Software Dependencies No The paper mentions implementing EMEF with "Pytorch" but does not specify a version number for PyTorch or any other software dependency.
Experiment Setup Yes In our experiments, λ is set to 0.002. The network architecture of the generator follows the image-to-image translation network (Isola et al. 2017). The image size of both input and output are 512 512. In the imitator network pre-training, the batch size is set to 1 and the network is trained with an Adam optimizer for 100 epochs. In the first 50 epochs, the learning rate is set to 2e 4, and then decays linearly for the rest.