arXiv subjects

Motahar Reza

Publications and source records attributed to Motahar Reza.

5 recordsLinked to original sources

arXiv2024

A Lightweight Transformer with Phase-Only Cross-Attention for Illumination-Invariant Biometric Authentication

Traditional biometric systems have encountered significant setbacks due to various unavoidable factors, for example, wearing of face masks in face recognition-based biometrics and hygiene concerns in fingerprint-based biometrics. This paper proposes a novel lightweight vision transformer with phase-only cross-attention (POC-ViT) using dual biometric traits of forehead and periocular portions of the face, capable of performing well even with face masks and without any physical touch, offering a promising alternative to traditional methods. The POC-ViT framework is designed to handle two biometric traits and to capture inter-dependencies in terms of relative structural patterns. Each channel consists of a Cross-Attention using phase-only correlation (POC) that captures both their individual and correlated structural patterns. The computation of cross-attention using POC extracts the phase correlation in the spatial features. Therefore, it is robust against variations in resolution and intensity, as well as illumination changes in the input images. The lightweight model is suitable for edge device deployment. The performance of the proposed framework was successfully demonstrated using the Forehead Subcutaneous Vein Pattern and Periocular Biometric Pattern (FSVP-PBP) database, having 350 subjects. The POC-ViT framework outperformed state-of-the-art methods with an outstanding classification accuracy of $98.8\%$ with the dual biometric traits.

cs.CV↗

arXiv2020

Using a Bi-directional LSTM Model with Attention Mechanism trained on MIDI Data for Generating Unique Music

Generating music is an interesting and challenging problem in the field of machine learning. Mimicking human creativity has been popular in recent years, especially in the field of computer vision and image processing. With the advent of GANs, it is possible to generate new similar images, based on trained data. But this cannot be done for music similarly, as music has an extra temporal dimension. So it is necessary to understand how music is represented in digital form. When building models that perform this generative task, the learning and generation part is done in some high-level representation such as MIDI (Musical Instrument Digital Interface) or scores. This paper proposes a bi-directional LSTM (Long short-term memory) model with attention mechanism capable of generating similar type of music based on MIDI data. The music generated by the model follows the theme/style of the music the model is trained on. Also, due to the nature of MIDI, the tempo, instrument, and other parameters can be defined, and changed, post generation.

cs.SD↗

arXiv2020

Graph based Clustering Algorithm for Social Community Transmission Prediction of COVID-19

A system to model the spread of COVID-19 cases after lockdown has been proposed, to define new preventive measures based on hotspots, using the graph clustering algorithm. This method allows for more lenient measures in areas less prone to the virus spread. There exist methods to model the spread of the virus, by predicting the number of confirmed cases. But the proposed system focuses more on the preventive side of the solution from a geographical point of view, by predicting the areas or regions that may become hotspots for the virus in the near future. The fact that the virus can only be transmitted by being in close proximity to an already infected person, suggests that, the regions that can easily be reached from an existing hotspot, have a higher chance of becoming a new hotspot. Moreover, in smaller regions, even after strict provisions, positive cases have been found. To consider this fact, the geographic distance between the nearest hotspots can be used as a measure of likelihood of the region also becoming a hotspot. In this paper, a weighted graph of regions with the regions themselves as weighted nodes with weight of the nodes as the number of active cases and the distance as edge weights. The graph can be completely connected or connected based on a distance threshold. The nodes are the administrative, and the distance measure tells the possible transmission between separate communities. Using this data, the potential regions that can become hotspots can be predicted, and preventive measures can be devised.

cs.SI↗

arXiv2020

A Parallel Approach for Real-Time Face Recognition from a Large Database

We present a new facial recognition system, capable of identifying a person, provided their likeness has been previously stored in the system, in real time. The system is based on storing and comparing facial embeddings of the subject, and identifying them later within a live video feed. This system is highly accurate, and is able to tag people with their ID in real time. It is able to do so, even when using a database containing thousands of facial embeddings, by using a parallelized searching technique. This makes the system quite fast and allows it to be highly scalable.

cs.CV↗

arXiv2019

Forced convection in a fluid saturated anisotropic porous channel with isoflux boundaries

Fully developed forced convective flow in a channel filled with porous material bounded by two impermeable walls subject to constant heat flux is considered. We consider Brinkman-Forchhimer equation to govern the flow inside the porous medium which accounts for the presence of inertial term. We assume that the porous medium is anisotropic in nature and permeability is varying along all the directions, correspondingly, permeability will be a matrix which appear as a positive semi-definite matrix in the momentum equation. The anisotropic nature of the porous medium is characterized by anisotropic ratio and angle. We have obtained the velocity, temperature and Nusselt number numerically, due to the presence of the non-linear quadratic term in the momentum equation. The effect of anisotropic permeability ratio and anisotropic angle on hydrodynamic and heat transfer is shown. We see that Nusselt number is altered significantly with the anisotropic ratio and angle. We present a fresh analysis about the inclusion of the permeability matrix in the Brinkman-Forchhimer extended Darcy momentum equation. To the best of the authors knowledge, this is the first attempt to analyze general anisotropy corresponding to Brinkman-Forchhimer extended Darcy model.

physics.flu-dyn↗