no code implementations • 5 Feb 2024 • Umut Cem Entok, Firas Laakom, Farhad Pakdaman, Moncef Gabbouj
Motivated by this, we propose a novel multi-illuminant color constancy method, by learning pixel-wise illumination maps caused by multiple light sources.
no code implementations • 5 Jan 2024 • Firas Laakom, Yuheng Bu, Moncef Gabbouj
Existing generalization theories of supervised learning typically take a holistic approach and provide bounds for the expected generalization over the whole data distribution, which implicitly assumes that the model generalizes similarly for all the classes.
no code implementations • 25 Sep 2023 • Firas Laakom, Fahad Sohrab, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj
We show that such regularizers improve performance.
no code implementations • 25 Sep 2023 • Fahad Sohrab, Firas Laakom, Moncef Gabbouj
The objective of S-SVDD is to map the original data to a subspace optimized for one-class classification, and the iterative optimization process of data mapping and description in S-SVDD relies on gradient descent.
no code implementations • ICLR Workshop EBM 2021 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj
Energy-based learning is a powerful learning paradigm that encapsulates various discriminative and generative approaches.
no code implementations • 3 Jan 2023 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj
At each optimization step, neurons at a given layer receive feedback from neurons belonging to higher layers of the hierarchy.
no code implementations • 22 Sep 2022 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj
In this paper, we propose an approach that builds on top of BoF pooling to boost its efficiency by ensuring that the items of the learned dictionary are non-redundant.
no code implementations • 9 Feb 2022 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj
We tested our approach across different tasks: dimensionality reduction using three different dataset, image compression using the MNIST dataset, and image denoising using fashion MNIST.
no code implementations • 9 Feb 2022 • Firas Laakom, Jenni Raitoharju, Nikolaos Passalis, Alexandros Iosifidis, Moncef Gabbouj
In this paper, we consider the problem of non-linear dimensionality reduction under uncertainty, both from a theoretical and algorithmic perspectives.
1 code implementation • 10 Nov 2021 • Firas Laakom, Jenni Raitoharju, Jarno Nikkanen, Alexandros Iosifidis, Moncef Gabbouj
We test this approach on the proposed method and show that it can indeed be used to avoid several extreme error cases and, thus, improves the practicality of the proposed technique.
1 code implementation • 10 Nov 2021 • Firas Laakom, Kateryna Chumachenko, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj
Based on this idea, we propose to reformulate the attention mechanism in CNNs to learn to ignore instead of learning to attend.
no code implementations • 29 Sep 2021 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj
Neural networks are composed of multiple layers arranged in a hierarchical structure jointly trained with a gradient-based optimization, where the errors are back-propagated from the last layer back to the first one.
no code implementations • 10 Jun 2021 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj
We study the diversity of the features learned by a two-layer neural network trained with the least squares loss.
no code implementations • 1 Jan 2021 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj
During the last decade, neural networks have been intensively used to tackle various problems and they have often led to state-of-the-art results.
no code implementations • 1 Sep 2020 • Firas Laakom, Jenni Raitoharju, Nikolaos Passalis, Alexandros Iosifidis, Moncef Gabbouj
spectral-based subspace learning is a common data preprocessing step in many machine learning pipelines.
no code implementations • 20 Jul 2020 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Jarno Nikkanen, Moncef Gabbouj
Computational color constancy is a preprocessing step used in many camera systems.
no code implementations • 6 May 2020 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Uygar Tuna, Jarno Nikkanen, Moncef Gabbouj
In this paper, we propose a novel unsupervised color constancy method, called Probabilistic Color Constancy (PCC).
1 code implementation • 23 Oct 2019 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Jarno Nikkanen, Moncef Gabbouj
In this paper, we describe a new large dataset for illumination estimation.
Few-Shot Camera-Adaptive Color Constancy Image Declipping +1
1 code implementation • 11 Jun 2019 • Firas Laakom, Nikolaos Passalis, Jenni Raitoharju, Jarno Nikkanen, Anastasios Tefas, Alexandros Iosifidis, Moncef Gabbouj
To further improve the illumination estimation accuracy, we propose a novel attention mechanism for the BoCF model with two variants based on self-attention.
no code implementations • 4 Jun 2019 • Firas Laakom, Jenni Raitoharju, Alexandros Iosifidis, Jarno Nikkanen, Moncef Gabbouj
In this paper, we study the importance of pre-training for the generalization capability in the color constancy problem.