• Corpus ID: 243938726

Robust channel-wise illumination estimation

@inproceedings{Laakom2021RobustCI,
  title={Robust channel-wise illumination estimation},
  author={Firas Laakom and Jenni Raitoharju and Jarno Nikkanen and Alexandros Iosifidis and M. Gabbouj},
  booktitle={BMVC},
  year={2021}
}
Recently, Convolutional Neural Networks (CNNs) have been widely used to solve the illuminant estimation problem and have often led to state-of-the-art results. Standard approaches operate directly on the input image. In this paper, we argue that this problem can be decomposed into three channel-wise independent and symmetric sub-problems and propose a novel CNN-based illumination estimation approach based on this decomposition. The proposed method substantially reduces the number of parameters… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 45 REFERENCES

Monte Carlo Dropout Ensembles for Robust Illumination Estimation

This paper proposes to aggregate different deep learning methods according to their output uncertainty using Monte Carlo dropout and obtains the final illumination estimate as the sum of the different model estimates weighted by the log-inverse of their corresponding uncertainties.

Color constancy using CNNs

Taking image patches as input, the CNN works in the spatial domain without using hand-crafted features that are employed by most previous methods to accurately predict the scene illumination.

Deep Outdoor Illumination Estimation

It is demonstrated that the approach allows the recovery of plausible illumination conditions and enables photorealistic virtual object insertion from a single image and significantly outperforms previous solutions to this problem.

FC^4: Fully Convolutional Color Constancy with Confidence-Weighted Pooling

This work presents a fully convolutional network architecture in which patches throughout an image can carry different confidence weights according to the value they provide for color constancy estimation, which allows for end-to-end training, and achieves higher efficiency and accuracy.

Color Constancy by Deep Learning

A framework using Deep Neural Networks (DNNs) to obtain an accurate light source estimator to achieve color constancy as a DNN-based regression approach to estimate the color of the light source.

Color Constancy Convolutional Autoencoder

Two novel approaches based on convolutional autoencoders are proposed: an unsupervised pre-training algorithm using a fine-tuned encoder and a semi-supervisedPre-training algorithms using a novel composite-loss function, which enables to solve the data scarcity problem and achieve competitive results while requiring much fewer parameters on ColorChecker RECommended dataset.

Unsupervised Learning for Color Constancy

An unsupervised learning-based method is proposed that learns its parameter values after approximating the unknown ground-truth illumination of the training images, thus avoiding calibration and outperforms all statistics-based and many learning- based methods in terms of accuracy.

Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review

This review, which focuses on the application of CNNs to image classification tasks, covers their development, from their predecessors up to recent state-of-the-art deep learning systems.

A Multi-Hypothesis Approach to Color Constancy

This work proposes a Bayesian framework that naturally handles color constancy ambiguity via a multi-hypothesis strategy and provides state-of-the-art accuracy on multiple public datasets (up to 11% median angular error improvement) while maintaining real-time execution.

Scene Understanding in Deep Learning-Based End-to-End Controllers for Autonomous Vehicles

Two complementary frameworks to automatically determine the most contributive regions of the input scenes are proposed, offering intuitive knowledge of how a trained end-to-end autonomous vehicle controller understands driving scenarios.