Nikolaos Mitianoudis

Scopus Publications

Interpretable Vision Transformers in Monocular Depth Estimation via SVDA
Vasileios Arampatzakis, George Pavlidis, Nikolaos Mitianoudis, Nikos Papamarkos
Mathematics, 2026
Monocular depth estimation is a central problem in computer vision with applications in robotics, augmented reality, and autonomous driving, yet the self-attention mechanisms used by modern Transformer architectures remain opaque. In this work, we integrate SVD-Inspired Attention (SVDA) into the Dense Prediction Transformer (DPT), introducing a spectrally structured attention formulation for dense prediction that decouples directional alignment from spectral modulation through a learnable diagonal matrix embedded in normalized query–key interactions. Experiments on KITTI and NYU-v2 show that SVDA preserves competitive predictive performance while enabling intrinsic interpretability: on KITTI, AbsRel improves from 0.058 to 0.056 and δ1 from 0.976 to 0.979, while on NYU-v2, AbsRel improves from 0.133 to 0.124 and δ1 from 0.865 to 0.872. This is achieved with only 0.01% additional parameters, at the cost of a measurable runtime overhead associated with the added normalization and spectral modulation. More importantly, SVDA enables six spectral indicators that quantify entropy, rank, sparsity, alignment, selectivity, and robustness, revealing consistent cross-dataset and depth-wise patterns in how attention organizes during training. These properties make the model easier to inspect and better suited to applications where transparency and reliability are important, such as robotics and autonomous navigation.
On Segment-Aware Monocular Depth Estimation Using Vision Transformers
Vasileios Arampatzakis, George Pavlidis, Nikolaos Mitianoudis, Nikos Papamarkos
Information Switzerland, 2026
Monocular Depth Estimation (MDE) infers per-pixel scene geometry from a single RGB image. Despite recent progress, global MDE models often blur depth discontinuities at object boundaries and fail to capture object-level structure. Segment-aware depth estimation addresses this limitation by exploiting semantic segmentation to decompose depth prediction into simpler, class-specific subproblems. In this work, we study semantic-aware MDE in a multi-branch design where each semantic class is handled by a lightweight Vision Transformer (ViT) branch that predicts dense depth for its class while suppressing interference from other regions. We further examine fusion strategies that merge the branch outputs into a single prediction: (i) a learnable cross-attention fusion module that predicts depth from the stack of per-class proposals and masks, and (ii) a parameter-free stitched summation that sums mask-gated outputs. The proposed architecture is simple, scalable, end-to-end trainable, and compatible with arbitrary transformer backbones. Experiments on Virtual KITTI 2, where ground-truth depth and semantic labels are available, show that segment-aware modeling produces sharper depth boundaries and improves standard error metrics compared to a single-branch baseline (AbsRel 0.243→0.152; RMSE 11.952→9.101). Finally, we find that the parameter-free summation matches, and in most cases improves upon, the accuracy of learned fusion while adding no computational overhead.
Interpretable Vision Transformers in Image Classification via SVDA
Vasileios Arampatzakis, George Pavlidis, Nikolaos Mitianoudis, Nikos Papamarkos
IEEE Access, 2026
Vision Transformers (ViTs) have achieved state-of-the-art performance in image classification, yet their attention mechanisms often remain opaque and exhibit dense, non-structured behaviors. In this work, we adapt our previously proposed SVD-Inspired Attention (SVDA) mechanism to the ViT architecture, introducing a geometrically grounded formulation that enhances interpretability, sparsity, and spectral structure. We apply the use of interpretability indicators—originally proposed with SVDA—to monitor attention dynamics during training and assess structural properties of the learned representations. Experimental evaluations on four widely used benchmarks—CIFAR-10, FashionMNIST, CIFAR-100, and ImageNet-100—together with an additional pretrained fine-tuning study in a standard ViT setting show that SVDA preserves competitive classification behavior in our experimental settings while providing descriptive diagnostics of attention structure. In the pretrained setting, we integrate the exact SVDA operator into the late transformer blocks of a standard pretrained ViT and fine-tune on ImageNet-100, providing additional evidence that the proposed mechanism remains viable beyond compact from-scratch training. While the current framework offers descriptive insights rather than prescriptive guidance, our results establish SVDA as a comprehensive and informative tool for analyzing and developing structured attention models in computer vision. This work lays the foundation for future advances in explainable AI, spectral diagnostics, and attention-based model design.
Geometry Meets Attention: Interpretable Transformers via SVD Inspiration
Vasileios Arampatzakis, George Pavlidis, Nikolaos Mitianoudis, Nikos Papamarkos
IEEE Access, 2025
Self-attention is a cornerstone of modern deep learning, yet its dense dot-product formulation offers limited interpretability and lacks explicit structural constraints. We propose <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SVD-inspired Attention</i> (SVDA), a novel self-attention mechanism that introduces normalized query/key projections and a learnable diagonal spectral modulation, drawing direct motivation from the structure of Singular Value Decomposition (SVD). This formulation separates directional alignment from spectral emphasis, offering a geometrically grounded and interpretable variant of attention. We formalize SVDA within a standard multi-head Transformer architecture and introduce a suite of structure-aware indicators—such as spectral entropy, effective rank, and selectivity—that quantify interpretability and sparsity in attention dynamics. Our analysis highlights SVDA’s capacity for structured, energy-aware attention without compromising architectural compatibility or expressiveness. This work provides a theoretical foundation and diagnostic framework for structured attention models aimed at interpretability, compression, and semantic transparency.
Towards Explainability in Monocular Depth Estimation
Vasileios Arampatzakis, George Pavlidis, Kyriakos Pantoglou, Nikolaos Mitianoudis, Nikos Papamarkos
Communications in Computer and Information Science, 2025
Person Identification Using Temporal Analysis of Facial Blood Flow
Maria Raia, Thomas Stogiannopoulos, Nikolaos Mitianoudis, Nikolaos V. Boulgouris
Electronics Switzerland, 2024
Biometrics play an important role in modern access control and security systems. The need of novel biometrics to complement traditional biometrics has been at the forefront of research. The Facial Blood Flow (FBF) biometric trait, recently proposed by our team, is a spatio-temporal representation of facial blood flow, constructed using motion magnification from facial areas where skin is visible. Due to its design and construction, the FBF does not need information from the eyes, nose, or mouth, and, therefore, it yields a versatile biometric of great potential. In this work, we evaluate the effectiveness of novel temporal partitioning and Fast Fourier Transform-based features that capture the temporal evolution of facial blood flow. These new features, along with a “time-distributed” Convolutional Neural Network-based deep learning architecture, are experimentally shown to increase the performance of FBF-based person identification compared to our previous efforts. This study provides further evidence of FBF’s potential for use in biometric identification.
MResTNet: A Multi-Resolution Transformer Framework with CNN Extensions for Semantic Segmentation
Nikolaos Detsikas, Nikolaos Mitianoudis, Ioannis Pratikakis
Journal of Imaging, 2024
A fundamental task in computer vision is the process of differentiation and identification of different objects or entities in a visual scene using semantic segmentation methods. The advancement of transformer networks has surpassed traditional convolutional neural network (CNN) architectures in terms of segmentation performance. The continuous pursuit of optimal performance, with respect to the popular evaluation metric results, has led to very large architectures that require a significant amount of computational power to operate, making them prohibitive for real-time applications, including autonomous driving. In this paper, we propose a model that leverages a visual transformer encoder with a parallel twin decoder, consisting of a visual transformer decoder and a CNN decoder with multi-resolution connections working in parallel. The two decoders are merged with the aid of two trainable CNN blocks, the fuser that combined the information from the two decoders and the scaler that scales the contribution of each decoder. The proposed model achieves state-of-the-art performance on the Cityscapes and ADE20K datasets, maintaining a low-complexity network that can be used in real-time applications.
Automatic processing of dynamic responses via wavelet transform for the development of dynamic equivalent models
Efthimios P. Saroudis, Theofilos A. Papadopoulos, Georgios A. Barzegkar-Ntovom, Eleftherios O. Kontis, Nikolaos Mitianoudis, Grigoris K. Papagiannis
Sustainable Energy Grids and Networks, 2024
Monocular Depth Estimation: A Thorough Review
Vasileios Arampatzakis, George Pavlidis, Nikolaos Mitianoudis, Nikos Papamarkos
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024
Estimation of depth in two-dimensional images is among the challenging topics in Computer Vision. This is a well-studied but also an ill-posed problem, which has long been the focus of intense research. This paper is an in-depth review of the topic, presenting two aspects, one that considers the mechanisms of human depth perception, and another that includes the various Deep Learning approaches. The methods are presented in a compact and structured way that outlines the topic and categorizes the approaches according to the line of research followed in the recent decade. Although there has been significant advancement in the topic, it was without any connection with human depth perception and the potential benefits from this sector.
Non-Contact Blood Pressure Estimation Using Forehead and Palm Infrared Video
Thomas Stogiannopoulos, Nikolaos Mitianoudis
Biomedinformatics, 2024
This study investigates the potential of low-cost infrared cameras for non-contact monitoring of blood pressure (BP) in individuals with fragile health, particularly the elderly. Previous research has shown success in developing non-contact BP monitoring using RGB cameras. In this study, the Eulerian Video Magnification (EVM) technique is employed to enhance minor variations in skin pixel intensity in specific facial regions captured by an infrared camera from the forehead and palm. The primary focus of this study is to explore the possibility of using infrared cameras for non-contact BP monitoring under low-light or night-time conditions. We have successfully shown that by employing a series of straightforward signal processing techniques and regression analysis, we were able to achieve commendable outcomes in our experimental setup. Specifically, we were able to surpass the stringent accuracy standards set forth by the British Hypertension Society (BHS) and the Association for the Advancement of Medical Instrumentation (AAMI) protocol.
A Dilated MultiRes Visual Attention U-Net for historical document image binarization
Nikolaos Detsikas, Nikolaos Mitianoudis, Nikolaos Papamarkos
Signal Processing Image Communication, 2024
Sound Event Detection in Domestic Environment Using Frequency-Dynamic Convolution and Local Attention
Grigorios-Aris Cheimariotis, Nikolaos Mitianoudis
Information Switzerland, 2023
A Study of Machine Learning Regression Techniques for Non-Contact SpO2 Estimation from Infrared Motion-Magnified Facial Video
Thomas Stogiannopoulos, Grigorios-Aris Cheimariotis, Nikolaos Mitianoudis
Information Switzerland, 2023
Historical Document Image Binarization Using a Lightweight U-Net Derivative Architecture
Nikolaos Detsikas, Nikolaos Mitianoudis
International Workshop on Cellular Nanoscale Networks and their Applications, 2023
A lightweight ConvGRU network for Distracted Driving detection
Pantazis Anagnostou, Nikolaos Mitianoudis
International Conference on Digital Signal Processing DSP, 2023
Contactless Blood Pressure Estimation Using Infrared Video from Facial and Hand Regions
Thomas Stogiannopoulos, Nikolaos Mitianoudis
International Workshop on Cellular Nanoscale Networks and their Applications, 2023
A Convolutional Neural Network-Based Conditional Random Field Model for Structured Multi-Focus Image Fusion Robust to Noise
Odysseas Bouzos, Ioannis Andreadis, Nikolaos Mitianoudis
IEEE Transactions on Image Processing, 2023
A non-contact SpO2 estimation using video magnification and infrared data
Thomas Stogiannopoulos, Grigorios-Aris Cheimariotis, Nikolaos Mitianoudis
ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings, 2023
OptiNET—Automatic Network Topology Optimization
Andreas Maniatopoulos, Paraskevi Alvanaki, Nikolaos Mitianoudis
Information Switzerland, 2022
Conditional Random Field-Guided Multi-Focus Image Fusion
Odysseas Bouzos, Ioannis Andreadis, Nikolaos Mitianoudis
Journal of Imaging, 2022
Low-Cost Online Convolution Checksum Checker
Dionysios Filippas, Nikolaos Margomenos, Nikolaos Mitianoudis, Chrysostomos Nicopoulos, Giorgos Dimitrakopoulos
IEEE Transactions on Very Large Scale Integration VLSI Systems, 2022
Technical analysis forecasting and evaluation of stock markets: the probabilistic recovery neural network approach
Andreas Maniatopoulos, Alexandros Gazis, Nikolaos Mitianoudis
International Journal of Economics and Business Research, 2022
An Efficient Short-Time Discrete Cosine Transform and Attentive MultiResUNet Framework for Music Source Separation
Thomas Sgouros, Angelos Bousis, Nikolaos Mitianoudis
IEEE Access, 2022
Learnable Leaky ReLU (LeLeLU): An Alternative Accuracy-Optimized Activation Function
Andreas Maniatopoulos, Nikolaos Mitianoudis
Information Switzerland, 2021
Deep Person Identification Using Spatiotemporal Facial Motion Amplification
K. Gkentsidis, T. Pistola, N. Mitianoudis, N. V. Boulgouris
Proceedings International Conference on Image Processing Icip, 2020
A novel Directional Framework for Source Counting and Source Separation in Instantaneous Underdetermined Audio Mixtures
Thomas Sgouros, Nikolaos Mitianoudis
IEEE ACM Transactions on Audio Speech and Language Processing, 2020
Artificial Neural Network Performance Boost using Probabilistic Recovery with Fast Cascade Training
Andreas Maniatopoulos, Alexandros Gazis, Venetis P. Pallikaras, Nikolaos Mitianoudis
International Journal of Circuits Systems and Signal Processing, 2020
A Novel Scheme for Single-Channel Speech Dereverberation
Nikolaos Kilis, Nikolaos Mitianoudis
Acoustics, 2019
Biometric Identification Using Facial Motion Amplification
T. Pistola, A. Papadopoulos, N. Mitianoudis, N. V. Boulgouris
Proceedings International Conference on Image Processing Icip, 2019
Multiple-exposure image fusion for HDR image synthesis using learned analysis transformations †
Ioannis Merianos, Nikolaos Mitianoudis
Journal of Imaging, 2019
Conditional random field model for robust multi-focus image fusion
Odysseas Bouzos, Ioannis Andreadis, Nikolaos Mitianoudis
IEEE Transactions on Image Processing, 2019
Convolutive audio source separation using robust ICA and an intelligent evolving permutation ambiguity solution
Dimitrios Mallis, Thomas Sgouros, Nikolaos Mitianoudis
Evolving Systems, 2018
A Hermite neural network incorporating artificial bee colony optimization to model shoreline realignment at a reef-fronted beach
George E. Tsekouras, Vasilis Trygonis, Andreas Maniatopoulos, Anastasios Rigos, Antonios Chatzipavlis, John Tsimikas, Nikolaos Mitianoudis, Adonis F. Velegrakis
Neurocomputing, 2018
Multidimensional directional steerable filters — Theory and application to 3D flow estimation
Dimitrios S. Alexiadis, Nikolaos Mitianoudis, Tania Stathaki
Image and Vision Computing, 2018
Frequency-domain joint motion and disparity estimation using steerable filters
Dimitrios Alexiadis, Nikolaos Mitianoudis, Tania Stathaki
Inventions, 2018
Salt-n-pepper noise filtering using cellular automata
Journal of Cellular Automata, 2018
Stock prices predictability at long-horizons: Two tales from the time-frequency domain
Nikolaos Mitianoudis, Theologos Dergiades
Credit and Capital Markets, 2017
Underdetermined source separation using a sparse STFT framework and weighted Laplacian directional modelling
Thomas Sgouros, Nikolaos Mitianoudis
European Signal Processing Conference, 2016
A hybrid multiple exposure image fusion approach for HDR image synthesis
Ioannis Merianos, Nikolaos Mitianoudis
Ist 2016 2016 IEEE International Conference on Imaging Systems and Techniques Proceedings, 2016
Applying conformal geometry for creating a 3D model spatial-consistent texture map
George Ioannakis, Christodoulos Chamzas, Anestis Koutsoudis, Nikolaos Papamarkos, Ioannis Pratikakis, Fotis Arnaoutoglou, Nikolaos Mitianoudis, Thomas Sgouros
2016 Digital Media Industry and Academic Forum Dmiaf 2016 Proceedings, 2016
A memristive circular buffer for real-time signal processing
Christos Sichonidis, Ioannis Vourkas, Nikolaos Mitianoudis, Georgios Ch. Sirakoulis
2016 5th International Conference on Modern Circuits and Systems Technologies Mocast 2016, 2016
Converting a plant to a battery and wireless sensor with scatter radio and ultra-low cost
Christos Konstantopoulos, Eftichios Koutroulis, Nikolaos Mitianoudis, Aggelos Bletsas
IEEE Transactions on Instrumentation and Measurement, 2016
Convolutive audio source separation using robust ICA and reduced likelihood ratio jump
Dimitrios Mallis, Thomas Sgouros, Nikolaos Mitianoudis
IFIP Advances in Information and Communication Technology, 2016
Document image binarization using local features and Gaussian mixture modeling
Nikolaos Mitianoudis, Nikolaos Papamarkos
Image and Vision Computing, 2015
Local Co-occurrence and Contrast Mapping for Document Image Binarization
Nikolaos Mitianoudis, Nikolaos Papamarkos
Proceedings of International Conference on Frontiers in Handwriting Recognition Icfhr, 2014
Multi-spectral document image binarization using image fusion and background subtraction techniques
Nikolaos Mitianoudis, Nikolaos Papamarkos
2014 IEEE International Conference on Image Processing Icip 2014, 2014
Multidimensional steerable filters and 3D flow estimation
Dimitrios Alexiadis, Nikolaos Mitianoudis, Tania Stathaki
2014 IEEE International Conference on Image Processing Icip 2014, 2014
Real time hand detection in a complex background
Ekaterini Stergiopoulou, Kyriakos Sgouropoulos, Nikos Nikolaou, Nikos Papamarkos, Nikos Mitianoudis
Engineering Applications of Artificial Intelligence, 2014
MASTERS: A virtual lab on multimedia systems for telecommunications, medical, and remote sensing applications
Dimitrios S. Alexiadis, Nikolaos Mitianoudis
IEEE Transactions on Education, 2013
PCA summarization for audio song identification using Gaussian mixture models
Vaia Panagiotou, Nikolaos Mitianoudis
2013 18th International Conference on Digital Signal Processing DSP 2013, 2013
Region-based ICA image fusion using textural information
Nikolaos Mitianoudis, Sotirios-Antonios Antonopoulos, Tania Stathaki
2013 18th International Conference on Digital Signal Processing DSP 2013, 2013
A generalized directional Laplacian distribution: Estimation, mixture models and audio source separation
Nikolaos Mitianoudis
IEEE Transactions on Audio Speech and Language Processing, 2012
Region-based image fusion using a combinatory Chebyshev-ICA method
Zaid Omar, Nikolaos Mitianoudis, Tania Stathaki
ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings, 2011
Two-dimensional chebyshev polynomials for image fusion
Zaid Omar, Nikolaos Mitianoudis, Tania Stathaki
28th Picture Coding Symposium Pcs 2010, 2010
Fast wavelet-based pansharpening of multi-spectral images
Nikolaos Mitianoudis, Georgios Tzimiropoulos, Tania Stathaki
2010 IEEE International Conference on Imaging Systems and Techniques Ist 2010 Proceedings, 2010
A directional laplacian density for underdetermined audio source separation
Nikolaos Mitianoudis
Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics, 2010
A unifying approach to moment-based shape orientation and symmetry classification
G. Tzimiropoulos, N. Mitianoudis, T. Stathaki
IEEE Transactions on Image Processing, 2009
Optimal contrast for color image fusion using ICA bases
Proceedings of the 11th International Conference on Information Fusion Fusion 2008, 2008
Optimal Contrast Correction for ICA-Based Fusion of Multimodal Images
Nikolaos Mitianoudis, Tania Stathaki
IEEE Sensors Journal, 2008
Joint fusion and blind restoration for multiple image scenarios with missing data
N. Mitianoudis, T. Stathaki
Computer Journal, 2007
Applied multi-dimensional fusion
A. Mahmood, P. M. Tudor, W. Oxford, R. Hansford, J. D. B. Nelson, N. G. Kingsbury, A. Katartzis, M. Petrou, N. Mitianoudis, T. Stathaki, A. Achim, D. Bull, N. Canagarajah, S. Nikolov, A. Loza, N. Cvejic
Computer Journal, 2007
Robust recognition of planar shapes under affine transforms using principal component analysis
Georgios Tzimiropoulos, Nikolaos Mitianoudis, Tania Stathaki
IEEE Signal Processing Letters, 2007
An affine invariant function using PCA bases with an application to within-class object recognition
Georgios Tzimiropoulos, Nikolaos Mitianoudis, Tania Stathaki
ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings, 2007
Batch and online underdetermined source separation using Laplacian mixture models
Nikolaos Mitianoudis, Tania Stathaki
IEEE Transactions on Audio Speech and Language Processing, 2007
Spatial kernel K-harmonic means clustering for multi-spectral image segmentation
Q. Li, N. Mitianoudis, T. Stathaki
Iet Image Processing, 2007
Smooth signal extraction from instantaneous mixtures
Nikolaos Mitianoudis, Tania Stathaki, Anthony G. Constantinides
IEEE Signal Processing Letters, 2007
Underdetermined source separation using mixtures of warped Laplacians
Nikolaos Mitianoudis, Tania Stathaki
Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics, 2007
Pixel-based and region-based image fusion schemes using ICA bases
Nikolaos Mitianoudis, Tania Stathaki
Information Fusion, 2007
Compact representations of market securities using smooth component extraction
Hariton Korizis, Nikolaos Mitianoudis, Anthony G. Constantinides
Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics, 2007
Adaptive image fusion using ICA bases
ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings, 2006
Blind separation of skewed signals in instantaneous mixtures
N. Mitianoudis, T. Stathaki, M. Davies
IEEE Workshop on Signal Processing Systems Sips Design and Implementation, 2005
Overcomplete source separation using Laplacian mixture models
N. Mitianoudis, T. Stathaki
IEEE Signal Processing Letters, 2005
Audio source separation: Solutions and problems
Nikolaos Mitianoudis, Mike E. Davies
International Journal of Adaptive Control and Signal Processing, 2004
Simple mixture model for sparse overcomplete ICA
M. Davies, N. Mitianoudis
IEE Proceedings Vision Image and Signal Processing, 2004
Permutation alignment for frequency domain ICA using subspace beamforming methods
Nikolaos Mitianoudis, Mike Davies
Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics, 2004
A MAXIMUM LIKELIHOOD APPROACH TO BLIND AUDIO DE-REVERBERATION
Proceedings of the International Conference on Digital Audio Effects Dafx, 2004
Audio source separation of convolutive mixtures
N. Mitianoudis, M.E. Davies
IEEE Transactions on Speech and Audio Processing, 2003
Using beamforming in the audio source separation problem
N. Mitianoudis, M.E. Davies
Proceedings 7th International Symposium on Signal Processing and Its Applications Isspa 2003, 2003
A fixed point solution for convolved audio source separation
IEEE ASSP Workshop on Applications of Signal Processing to Audio and Acoustics, 2001

Nikolaos Mitianoudis

Scopus Publications

RECENT SCHOLAR PUBLICATIONS

MOST CITED SCHOLAR PUBLICATIONS