arXiv subjects

Hamidreza Amindavar

Publications and source records attributed to Hamidreza Amindavar.

12 recordsLinked to original sources

arXiv2026

CFGPNet: Cross-Attention-Based Fused Gradient Programmed Network Framework for Multispectral Object Detection

Multispectral object detection combines visible and thermal imagery to improve perception under challenging illumination and environmental conditions. However, differences in modality appearance and reliability can introduce redundant or conflicting responses, limiting the use of complementary information. Complex fusion mechanisms further increase computational cost, creating a persistent trade-off between detection accuracy and efficiency. To address these challenges, CFGPNet is proposed, a cross-attention-based fused gradient programmed network. The framework incorporates re-parameterized RepViT blocks into the YOLOv9 architecture to strengthen spatial and channel representations while maintaining efficient feature extraction. Cross Computation Efficient Attention (CrossCEA) exchanges spatial attention maps between modalities at multiple detection scales, allowing each stream to emphasize regions supported by the other while preserving modality-specific information. Attention Selection and Aggregation Fusion (ASAF) combines dense feature aggregation with selection of the strongest responses from multiple attention branches to form compact, discriminative fused representations. A programmable gradient information pathway provides auxiliary supervision during training to improve feature learning. This pathway is removed after training, adding no parameters or operations at inference. Experiments on FLIR, M3FD, LLVIP, VEDAI, and MFAD demonstrate favorable accuracy-efficiency trade-offs across three model scales, with the smallest variant requiring 15.3 million parameters and 56.9 GFLOPs. The code is available at https://github.com/NimaHatami99/CFGPNet.

cs.CV↗

arXiv2024

Enhanced Depth Estimation and 3D Geometry Reconstruction using Bayesian Helmholtz Stereopsis with Belief Propagation

Helmholtz stereopsis is one the versatile techniques for 3D geometry reconstruction from 2D images of objects with unknown and arbitrary reflectance surfaces. HS eliminates the need for surface reflectance, a challenging parameter to measure, based on the Helmholtz reciprocity principle. Its Bayesian formulation using maximum a posteriori (MAP) probability approach has significantly improved reconstruction accuracy of HS method. This framework enables the inclusion of smoothness priors which enforces observations and neighborhood information in the formulation. We used Markov Random Fields (MRF) which is a powerful tool to integrate diverse prior contextual information and solved the MAP-MRF using belief propagation algorithm. We propose a new smoothness function utilizing the normal field integration method for refined depth estimation within the Bayesian framework. Utilizing three pairs of images with different viewpoints, our approach demonstrates superior depth label accuracy compared to conventional Bayesian methods. Experimental results indicate that our proposed method yields a better depth map with reduced RMS error, showcasing its efficacy in improving depth estimation within Helmholtz stereopsis.

eess.IV↗

arXiv2023

Stellar Karaoke: deep blind separation of terrestrial atmospheric effects out of stellar spectra by velocity whitening

We report a study exploring how the use of deep neural networks with astronomical Big Data may help us find and uncover new insights into underlying phenomena: through our experiments towards unsupervised knowledge extraction from astronomical Big Data we serendipitously found that deep convolutional autoencoders tend to reject telluric lines in stellar spectra. With further experiments we found that only when the spectra are in the barycentric frame does the network automatically identify the statistical independence between two components, stellar vs telluric, and rejects the latter. We exploit this finding and turn it into a proof-of-concept method for removal of the telluric lines from stellar spectra in a fully unsupervised fashion: we increase the inter-observation entropy of telluric absorption lines by imposing a random, virtual radial velocity to the observed spectrum. This technique results in a non-standard form of ``whitening'' in the atmospheric components of the spectrum, decorrelating them across multiple observations. We process more than 250,000 spectra from the High Accuracy Radial velocity Planetary Search (HARPS) and with qualitative and quantitative evaluations against a database of known telluric lines, show that most of the telluric lines are successfully rejected. Our approach, `Stellar Karaoke', has zero need for prior knowledge about parameters such as observation time, location, or the distribution of atmospheric molecules and processes each spectrum in milliseconds. We also train and test on Sloan Digital Sky Survey (SDSS) and see a significant performance drop due to the low resolution. We discuss directions for developing tools on top of the introduced method in the future.

astro-ph.SR↗

arXiv2023

Swin-Tempo: Temporal-Aware Lung Nodule Detection in CT Scans as Video Sequences Using Swin Transformer-Enhanced UNet

Lung cancer is highly lethal, emphasizing the critical need for early detection. However, identifying lung nodules poses significant challenges for radiologists, who rely heavily on their expertise for accurate diagnosis. To address this issue, computer-aided diagnosis (CAD) systems based on machine learning techniques have emerged to assist doctors in identifying lung nodules from computed tomography (CT) scans. Unfortunately, existing networks in this domain often suffer from computational complexity, leading to high rates of false negatives and false positives, limiting their effectiveness. To address these challenges, we present an innovative model that harnesses the strengths of both convolutional neural networks and vision transformers. Inspired by object detection in videos, we treat each 3D CT image as a video, individual slices as frames, and lung nodules as objects, enabling a time-series application. The primary objective of our work is to overcome hardware limitations during model training, allowing for efficient processing of 2D data while utilizing inter-slice information for accurate identification based on 3D image context. We validated the proposed network by applying a 10-fold cross-validation technique to the publicly available Lung Nodule Analysis 2016 dataset. Our proposed architecture achieves an average sensitivity criterion of 97.84% and a competition performance metrics (CPM) of 96.0% with few parameters. Comparative analysis with state-of-the-art advancements in lung nodule identification demonstrates the significant accuracy achieved by our proposed model.

eess.IV↗

arXiv2022

Deep Bayesian Active Learning, A Brief Survey on Recent Advances

Active learning frameworks offer efficient data annotation without remarkable accuracy degradation. In other words, active learning starts training the model with a small size of labeled data while exploring the space of unlabeled data in order to select most informative samples to be labeled. Generally speaking, representing the uncertainty is crucial in any active learning framework, however, deep learning methods are not capable of either representing or manipulating model uncertainty. On the other hand, from the real world application perspective, uncertainty representation is getting more and more attention in the machine learning community. Deep Bayesian active learning frameworks and generally any Bayesian active learning settings, provide practical consideration in the model which allows training with small data while representing the model uncertainty for further efficient training. In this paper, we briefly survey recent advances in Bayesian active learning and in particular deep Bayesian active learning frameworks.

cs.LG↗

arXiv2019

A New Framework For Spatial Modeling And Synthesis of Genome Sequence

This paper provides a framework in order to statistically model sequences from human genome, which is allowing a formulation to synthesize gene sequences. We start by converting the alphabetic sequence of genome to decimal sequence by Huffman coding. Then, this decimal sequence is decomposed by HP filter into two components, trend and cyclic. Next, a statistical modeling, ARIMA-GARCH, is implemented on trend component exhibiting heteroskedasticity, autoregressive integrated moving average (ARIMA) to capture the linear characteristics of the sequence and later, generalized autoregressive conditional heteroskedasticity (GARCH) is then appropriated for the statistical nonlinearity of genome sequence. This modeling approach synthesizes a given genome sequence regarding to its statistical features. Finally, the PDF of a given sequence is estimated using Gaussian mixture model and based on estimated PDF, we determine a new PDF presenting sequences that counteract statistically the original sequence. Our strategy is performed on several genes as well as HIV nucleotide sequence and corresponding results is presented.

q-bio.OT↗

arXiv2019

Graph-Based Method for Anomaly Prediction in Brain Network

Resting-state functional MRI (rs-fMRI) in functional neuroimaging techniques have improved in brain disorders, dysfunction studies via mapping the topology of the brain connections, i.e. connectopic mapping. Since, there are the slight differences between healthy and unhealthy brain regions and functions, investigation into the complex topology of functional and structural brain networks in human is a complicated task with the growth of evaluation criteria. Irregular graph deep learning applications have widely spread to understanding human cognitive functions that are linked to gene expression and related distributed spatial patterns, because the neuronal networks of the brain can hold dynamically a variety of brain solutions with different activity patterns and functional connectivity, these applications might also be involved with both node-centric and graph-centric tasks. In this paper, we performed a novel approach of individual generative model and high order graph analysis for the region of interest recognition areas of the brain which do not have a normal connection during applying certain tasks. Here, we proposed a high order framework of Graph Auto-Encoder (GAE) with a hypersphere distributer for functional data analysis in brain imaging studies that is underlying non-Euclidean structure in the learning of strong non-rigid graphs among large scale data. In addition, we distinguished the possible modes of correlations in abnormal brain connections. Our finding will show the degree of correlation between the affected regions and their simultaneous occurrence over time that can be used to diagnose brain diseases or revealing the ability of the nervous system to modify in brain topology at all angles, brain plasticity, according to input stimuli.

eess.IV↗

arXiv2019

A Novel Maneuvering Target Tracking Approach by Stochastic Volatility GARCH Model

In this paper, we introduce a new single model maneuvering target tracking approach using stochastic differential equation (SDE) based on GARCH volatility. The traditional input estimation (IE) techniques assume constant acceleration level which do not cover all the possible acceleration quintessence. In contrast, the multiple model (MM) algorithms that take care of some IE's shortcomings, are sensitive to the transition probability matrices. In this paper, an innovative model is proposed to overcome these drawbacks by using a new generalized dynamic modeling of acceleration and a Bayesian filter. We utilize SDE to model Markovian jump acceleration of a maneuvering target through GARCH process as the SDE volatility. In the proposed scheme, the original state and stochastic volatility (SV) are estimated simultaneously by a bootstrap particle filter (PF). We introduce the bootstrap resampling to obtain the statistical properties of a GARCH density. Due to the heavy-tailed nature of the GARCH distribution, the bootstrap PF is more effective in the presence of large errors that can occur in the state equation. We show analytically that the target tracking performance is improved by considering GARCH acceleration model. Finally, the effectiveness and capabilities of our proposed strategy (PF-AR-GARCH) are demonstrated and validated through simulation studies.

stat.AP↗

arXiv2017

Semi-Supervised Hierarchical Semantic Object Parsing

Models based on Convolutional Neural Networks (CNNs) have been proven very successful for semantic segmentation and object parsing that yield hierarchies of features. Our key insight is to build convolutional networks that take input of arbitrary size and produce object parsing output with efficient inference and learning. In this work, we focus on the task of instance segmentation and parsing which recognizes and localizes objects down to a pixel level base on deep CNN. Therefore, unlike some related work, a pixel cannot belong to multiple instances and parsing. Our model is based on a deep neural network trained for object masking that supervised with input image and follow incorporates a Conditional Random Field (CRF) with end-to-end trainable piecewise order potentials based on object parsing outputs. In each CRF unit we designed terms to capture the short range and long range dependencies from various neighbors. The accurate instance-level segmentation that our network produce is reflected by the considerable improvements obtained over previous work at high APr thresholds. We demonstrate the effectiveness of our model with extensive experiments on challenging dataset subset of PASCAL VOC2012.

cs.AI↗

arXiv2014

Sparse Estimation with Generalized Beta Mixture and the Horseshoe Prior

In this paper, the use of the Generalized Beta Mixture (GBM) and Horseshoe distributions as priors in the Bayesian Compressive Sensing framework is proposed. The distributions are considered in a two-layer hierarchical model, making the corresponding inference problem amenable to Expectation Maximization (EM). We present an explicit, algebraic EM-update rule for the models, yielding two fast and experimentally validated algorithms for signal recovery. Experimental results show that our algorithms outperform state-of-the-art methods on a wide range of sparsity levels and amplitudes in terms of reconstruction accuracy, convergence rate and sparsity. The largest improvement can be observed for sparse signals with high amplitudes.

cs.IT↗

arXiv2013

Cyclostationary Approach for Heart and Respiration Rates Monitoring with Body Movement Cancellation Using Radar Doppler System

Heart and respiration rate measurement using Doppler radar is a non-contact and non-obstructive way for remote thorough-clothing monitoring of vital signs. The modulated back-scattered radar signal in the presence of high noise and interference is non-stationary with hidden periodicities, which cannot be detected by ordinary Fourier analysis. In this paper we propose a cyclostationary approach for such signals and show that by using non-linear transformation and then Fourier analysis of the radar signal, the hidden periodicities can be accurately obtained. Numerical results show that the vital signs can be extracted as cyclic frequencies, independent of SNR and without any filtering or phase unwrapping.

physics.med-ph↗

arXiv2011

A Hybrid Global Minimization Scheme for Accurate Source Localization in Sensor Networks

We consider the localization problem of multiple wideband sources in a multi-path environment by coherently taking into account the attenuation characteristics and the time delays in the reception of the signal. Our proposed method leaves the space for unavailability of an accurate signal attenuation model in the environment by considering the model as an unknown function with reasonable prior assumptions about its functional space. Such approach is capable of enhancing the localization performance compared to only utilizing the signal attenuation information or the time delays. In this paper, the localization problem is modeled as a cost function in terms of the source locations, attenuation model parameters and the multi-path parameters. To globally perform the minimization, we propose a hybrid algorithm combining the differential evolution algorithm with the Levenberg-Marquardt algorithm. Besides the proposed combination of optimization schemes, supporting the technical details such as closed forms of cost function sensitivity matrices are provided. Finally, the validity of the proposed method is examined in several localization scenarios, taking into account the noise in the environment, the multi-path phenomenon and considering the sensors not being synchronized.

cs.NI↗