1 code implementation • 15 Feb 2024 • Mariia Drozdova, Vitaliy Kinakh, Omkar Bait, Olga Taran, Erica Lastufka, Miroslava Dessauges-Zavadsky, Taras Holotyak, Daniel Schaerer, Slava Voloshynovskiy
Current techniques, such as CLEAN and PyBDSF, often fail to detect faint sources, highlighting the need for more accurate methods.
no code implementations • 11 Nov 2023 • Guillaume Quétant, Yury Belousov, Vitaliy Kinakh, Slava Voloshynovskiy
We present a novel information-theoretic framework, termed as TURBO, designed to systematically analyse and generalise auto-encoding methods.
no code implementations • 28 Sep 2023 • Yury Belousov, Olga Taran, Vitaliy Kinakh, Slava Voloshynovskiy
Copy detection patterns (CDP) present an efficient technique for product protection against counterfeiting.
1 code implementation • 21 Mar 2023 • Vitaliy Kinakh, Mariia Drozdova, Slava Voloshynovskiy
We present a new method of self-supervised learning and knowledge distillation based on the multi-views and multi-representations (MV-MR).
Ranked #2 on Unsupervised Image Classification on STL-10
no code implementations • 14 Dec 2022 • Joakim Tutt, Olga Taran, Roman Chaban, Brian Pulfer, Yury Belousov, Taras Holotyak, Slava Voloshynovskiy
Nowadays, copy detection patterns (CDP) appear as a very promising anti-counterfeiting technology for physical object protection.
no code implementations • 28 Oct 2022 • Yury Belousov, Brian Pulfer, Roman Chaban, Joakim Tutt, Olga Taran, Taras Holotyak, Slava Voloshynovskiy
In this paper, we address the problem of modeling a printing-imaging channel built on a machine learning approach a. k. a.
no code implementations • 11 Oct 2022 • Roman Chaban, Olga Taran, Joakim Tutt, Yury Belousov, Brian Pulfer, Taras Holotyak, Slava Voloshynovskiy
Since digital off-set printing represents great flexibility in terms of product personalized in comparison with traditional off-set printing, it looks very interesting to address the above concerns for digital off-set printers that are used by several companies for the CDP protection of physical objects.
no code implementations • 29 Sep 2022 • Brian Pulfer, Yury Belousov, Joakim Tutt, Roman Chaban, Olga Taran, Taras Holotyak, Slava Voloshynovskiy
Systems based on classical supervised learning and digital templates assume knowledge of fake CDP at training time and cannot generalize to unseen types of fakes.
1 code implementation • 11 Jul 2022 • Behrooz Razeghi, Flavio P. Calmon, Deniz Gunduz, Slava Voloshynovskiy
In this work, we propose a general family of optimization problems, termed as complexity-leakage-utility bottleneck (CLUB) model, which (i) provides a unified theoretical framework that generalizes most of the state-of-the-art literature for the information-theoretic privacy models, (ii) establishes a new interpretation of the popular generative and discriminative models, (iii) constructs new insights to the generative compression models, and (iv) can be used in the fair generative models.
no code implementations • 23 Jun 2022 • Brian Pulfer, Roman Chaban, Yury Belousov, Joakim Tutt, Olga Taran, Taras Holotyak, Slava Voloshynovskiy
While Deep Learning (DL) can be used as a part of the authentication system, to the best of our knowledge, none of the previous works has studied the performance of a DL-based authentication system against ML-based attacks on CDP with 1x1 symbol size.
no code implementations • 4 Mar 2022 • Olga Taran, Joakim Tutt, Taras Holotyak, Roman Chaban, Slavi Bonev, Slava Voloshynovskiy
In the recent years, the copy detection patterns (CDP) attracted a lot of attention as a link between the physical and digital worlds, which is of great interest for the internet of things and brand protection applications.
no code implementations • 20 Dec 2021 • Guillaume Quétant, Mariia Drozdova, Vitaliy Kinakh, Tobias Golling, Slava Voloshynovskiy
We present Turbo-Sim, a generalised autoencoder framework derived from principles of information theory that can be used as a generative model.
1 code implementation • 17 Dec 2021 • Vitaliy Kinakh, Mariia Drozdova, Guillaume Quétant, Tobias Golling, Slava Voloshynovskiy
The InfoSCC-GAN architecture is based on an unsupervised contrastive encoder built on the InfoNCE paradigm, an attribute classifier and an EigenGAN generator.
1 code implementation • 15 Dec 2021 • Samuel Klein, John A. Raine, Sebastian Pina-Otey, Slava Voloshynovskiy, Tobias Golling
Normalizing flows are diffeomorphic, typically dimension-preserving, models trained using the likelihood of the model.
1 code implementation • 5 Oct 2021 • Roman Chaban, Olga Taran, Joakim Tutt, Taras Holotyak, Slavi Bonev, Slava Voloshynovskiy
Nowadays, the modern economy critically requires reliable yet cheap protection solutions against product counterfeiting for the mass market.
1 code implementation • 5 Jun 2021 • Amir Ahooye Atashin, Behrooz Razeghi, Deniz Gündüz, Slava Voloshynovskiy
We study the role of information complexity in privacy leakage about an attribute of an adversary's interest, which is not known a priori to the system designer.
no code implementations • 8 Feb 2021 • Behrooz Razeghi, Sohrab Ferdowsi, Dimche Kostadinov, Flavio. P. Calmon, Slava Voloshynovskiy
In this paper, we propose a framework for privacy-preserving approximate near neighbor search via stochastic sparsifying encoding.
1 code implementation • 4 Feb 2020 • Sohrab Ferdowsi, Behrooz Razeghi, Taras Holotyak, Flavio P. Calmon, Slava Voloshynovskiy
We propose a practical framework to address the problem of privacy-aware image sharing in large-scale setups.
no code implementations • 2 Dec 2019 • Slava Voloshynovskiy, Mouad Kondah, Shideh Rezaeifar, Olga Taran, Taras Holotyak, Danilo Jimenez Rezende
In particular, we present a new interpretation of VAE family based on the IB framework using a direct decomposition of mutual information terms and show some interesting connections to existing methods such as VAE [2; 3], beta-VAE [11], AAE [12], InfoVAE [5] and VAE/GAN [13].
1 code implementation • 13 Sep 2019 • Sohrab Ferdowsi, Maurits Diephuis, Shideh Rezaeifar, Slava Voloshynovskiy
We make a minimal, but very effective alteration to the VAE model.
no code implementations • 15 Jul 2019 • Behrooz Razeghi, Taras Stanko, Boris Škorić, Slava Voloshynovskiy
We investigate the privacy of two approaches to (biometric) template protection: Helper Data Systems and Sparse Ternary Coding with Ambiguization.
no code implementations • 14 May 2019 • Olga Taran, Shideh Rezaeifar, Taras Holotyak, Slava Voloshynovskiy
The robustness of the system is achieved by a specially designed key based randomization.
no code implementations • 8 May 2019 • Shideh Rezaeifar, Behrooz Razeghi, Olga Taran, Taras Holotyak, Slava Voloshynovskiy
In this paper, we address the problem of data reconstruction from privacy-protected templates, based on recent concept of sparse ternary coding with ambiguization (STCA).
1 code implementation • CVPR 2019 • Olga Taran, Shideh Rezaeifar, Taras Holotyak, Slava Voloshynovskiy
The vulnerability of machine learning systems to adversarial attacks questions their usage in many applications.
1 code implementation • 18 Mar 2019 • Olga Taran, Slavi Bonev, Slava Voloshynovskiy
In recent years, printable graphical codes have attracted a lot of attention enabling a link between the physical and digital worlds, which is of great interest for the IoT and brand protection applications.
no code implementations • 31 Jan 2019 • Dimche Kostadinov, Behrooz Razdehi, Slava Voloshynovskiy
In this paper, we introduce a novel concept for learning of the parameters in a neural network.
no code implementations • 30 Jan 2019 • Dimche Kostadinov, Behrooz Razeghi, Taras Holotyak, Slava Voloshynovskiy
We introduce a clustering principle that is based on evaluation of a parametric min-max measure for the discriminative prior.
1 code implementation • 10 Sep 2018 • Shideh Rezaeifar, Olga Taran, Slava Voloshynovskiy
We also introduce a criterion based on Kullback-Leibler divergence to reject doubtful examples.
3 code implementations • 5 Sep 2018 • Olga Taran, Shideh Rezaeifar, Slava Voloshynovskiy
The majority of the proposed existing adversarial attacks are based on the differentiability of the DNN cost function. Defence strategies are mostly based on machine learning and signal processing principles that either try to detect-reject or filter out the adversarial perturbations and completely neglect the classical cryptographic component in the defence.
no code implementations • 20 May 2018 • Dimche Kostadinov, Behrooz Razeghi, Sohrab Ferdowsi, Slava Voloshynovskiy
This paper presents a locally decoupled network parameter learning with local propagation.
no code implementations • ICLR 2018 • Dimche Kostadinov, Slava Voloshynovskiy
A novel measure related to the discriminative prior is proposed and defined on the support intersection for the transform representations.
no code implementations • 31 Oct 2017 • Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov
We present the multi-layer extension of the Sparse Ternary Codes (STC) for fast similarity search where we focus on the reconstruction of the database vectors from the ternary codes.
no code implementations • 29 Sep 2017 • Behrooz Razeghi, Slava Voloshynovskiy, Dimche Kostadinov, Olga Taran
The sparsifying transform and privacy amplification are not symmetric for the data owner and data user.
no code implementations • 7 Jul 2017 • Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov
A learning-based framework for representation of domain-specific images is proposed where joint compression and denoising can be done using a VQ-based multi-layer network.
no code implementations • 1 May 2017 • Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov
Furthermore, we also propose a general-purpose pre-processing for natural images which makes them suitable for such quantization.
no code implementations • 26 Jan 2017 • Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov, Taras Holotyak
This paper addresses the problem of Approximate Nearest Neighbor (ANN) search in pattern recognition where feature vectors in a database are encoded as compact codes in order to speed-up the similarity search in large-scale databases.