WBFlow: Few-shot White Balance for sRGB Images via Reversible Neural Flows

Authors: Chunxiao Li, Xuejing Kang, Anlong Ming

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments show that WBFlow achieves superior camera generalization and accuracy on three public datasets as well as our rendered multiple camera s RGB dataset.
Researcher Affiliation Academia Chunxiao Li , Xuejing Kang , Anlong Ming School of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications {chunxiaol, kangxuejing, mal}@bupt.edu.cn
Pseudocode No The paper describes the model architecture and processes but does not provide any explicitly labeled pseudocode or algorithm blocks.
Open Source Code Yes Our code is available at https://github.com/ChunxiaoLe/WBFlow.
Open Datasets Yes Following [Afifi and Brown, 2020a], we randomly selected 12000 s RGB images from the first fold of Set1 [Afifi et al., 2019a; Cheng et al., 2014] to train WBFlow.
Dataset Splits No The paper mentions training and testing on specific datasets (Set1, Set1-Test, Set2, Rendered Cube Dataset) but does not explicitly describe a separate validation set or detailed train/validation/test splits.
Hardware Specification No The paper does not provide any specific hardware details such as GPU or CPU models used for running experiments.
Software Dependencies No We implemented WBFlow on Pytorch with CUDA support and used the Adam [Kingma and Ba, 2014] with β1 = 0.9 and learning rate 10 4 to optimize it.
Experiment Setup Yes For the experiments with all training images, we trained WBFlow for 340000 iterations with batch size 4. While for few-shot experiments, we trained CT for 15000 iterations. We used the Adam [Kingma and Ba, 2014] with β1 = 0.9 and learning rate 10 4 to optimize it. We used color jittering, average blur, geometric rotation, and flipping to augment data. During testing, following [Afifi and Brown, 2020a], we resized all input images to a maximum dimension of 656 pixels and set a color mapping procedure to compute the final white-balanced s RGB images.