arXiv subjects

Akshay Agarwal

Publications and source records attributed to Akshay Agarwal.

At least 19 recordsLinked to original sources

arXiv2026

SegWave: Wavelet-Driven Segmentation of Tampered Regions

Verifying image authenticity is increasingly difficult, posing serious risks across journalism, law enforcement, and political domains. Most existing forensic methods rely on high-level visual artifacts and treat frame detection as a simple binary task. To address this, we propose SegWave, a hybrid framework that jointly leverages spatial and frequency-domain cues for image tampering detection. SegWave integrates a transformer-based architecture with the Discrete Wavelet Transform (DWT) to capture localized, multi-scale frequency inconsistencies indicative of manipulation. To further improve localization effectiveness, we introduce an Adaptive Sub-band Attention module (ASA) that dynamically highlights the informative high-frequency wavelet components. Extensive experiments on multiple benchmark datasets demonstrate that SegWave consistently outperforms state-of-the-art tampering detection methods in challenging evaluation settings.

cs.CV↗

arXiv2026

Evaluating Cross-lingual Knowledge Consistency in Code-Mixed vis-a-vis Indian Languages using IndicKLAR

Large language models recall knowledge reliably in English but often fail on the same query posed in a lower-resourced language -- a crosslingual consistency gap that remains underexplored for Indian languages and their code-mixed counterparts. To study this gap, we introduce IndiKLAR, an Indic extension of the KLAR-CLC benchmark covering 18 of the 22 scheduled Indian languages and pairing them with code-mixed variants for 11 widely used language pairs, with native-speaker verification of both monolingual and code-mixed variants for these 11 settings. This three-way alignment offers a unique opportunity to examine how knowledge recall consistency varies across the spectrum of English, code-mixed, and native Indian language inputs. Evaluating across nine open-weight models, we find that the native-language accuracy gap to English can reach $\sim$0.50, while code-mixed inputs close most of it -- bringing performance within $\sim$0.05 of English without any model-level intervention. Motivated by this, we evaluate several prompting strategies that vary in how language conversion is exposed, including a two-stage translate-then-answer setup, a one-stage joint translation-and-answer prompt, and Translate-in-Thought (TinT) -- a single-step strategy in which the model converts the input internally and emits only the final answer. Across the performance trajectory native $\rightarrow$ code-mixed $\rightarrow$ English, we identify a consistent flip point -- the boundary between incorrect and correct prediction -- that lies between the native and code-mixed settings. Interestingly, this holds whether the trajectory is induced by the input surface form or by the model's internal conversion process.

cs.CL↗

arXiv2026

The Unseen Adversaries: Robust and Generalized Defense Against Adversarial Patches

The vulnerabilities of deep neural networks against singularities have raised serious concerns regarding their deployment in the physical world. One of the most prominent and impactful physical-world adversarial perturbations is the attachment of patches to clean images, known as an adversarial patch attack. Similarly, natural noises such as Gaussian and Salt\&Pepper are highly prevalent in the real world. The current research need arises from the above vulnerabilities and the lack of efforts to tackle these two singularities independently and, especially, in combination. In this research, we have, for the first time, combined these two prominent singularities and proposed a novel dataset. Using this dataset, we have conducted a benchmark study of singularity data-point detection using features from several convolutional neural networks. For classification, rather than the popular neural network-based parameter tuning, we have used traditional yet effective machine learning classifiers. The extensive experiments across various in- and out-of-distribution (OOD) singularities reveal several interesting findings about the effectiveness of classifiers and show that it is hard to defend against adversaries when they are treated independently, and inefficient classifiers are selected.

cs.CV↗

arXiv2026

Beam Cross Sections Create Mixtures: Improving Feature Localization in Secondary Electron Imaging

Secondary electron (SE) imaging techniques, such as scanning electron microscopy and helium ion microscopy (HIM), use electrons emitted by a sample in response to a focused beam of charged particles incident at a grid of raster scan positions. Spot size -- the diameter of the incident beam's spatial profile -- is one of the limiting factors for resolution, along with various sources of noise in the SE signal. The effect of the beam spatial profile is commonly understood as convolutional. We show that under a simple and plausible physical abstraction for the beam, though convolution describes the mean of the SE counts, the full distribution of SE counts is a mixture. We demonstrate that this more detailed modeling can enable resolution improvements over conventional estimators through a stylized application inspired by semiconductor inspection: localizing the edge in a two-valued sample. We derive Fisher information about edge location in conventional and time-resolved measurements (TRM) and also derive the maximum likelihood estimate (MLE) from the latter. Empirically, the MLE computed from TRM is approximately efficient except at very low beam diameter, so Fisher information comparisons are predictive of performance and can be used to optimize the beam diameter relative to the raster scan spacing. Monte Carlo simulations provide an example of the MLE giving a 5-fold reduction in root mean-squared error (RMSE) of edge localization as compared to conventional interpolation-based estimation. The RMSE is substantially below both the beam diameter and the raster scan spacing and thus sub-pixel localization is demonstrated. Applied to three real HIM datasets, the average RMSE reduction factor is 5.4.

eess.SP↗

arXiv2026

Ozone Cues Mitigate Reflected Downwelling Radiance in LWIR Absorption-Based Ranging

Passive long-wave infrared (LWIR) absorption-based ranging relies on atmospheric absorption to estimate distances to objects from their emitted thermal radiation. First demonstrated decades ago for objects much hotter than the air and recently extended to scenes with low temperature variations, this ranging has depended on reflected radiance being negligible. Downwelling radiance is especially problematic, sometimes causing large inaccuracies. In two new ranging methods, we use characteristic features from ozone absorption to estimate the contribution of reflected downwelling radiance. The quadspectral method gives a simple closed-form range estimate from four narrowband measurements, two at a water vapor absorption line and two at an ozone absorption line. The hyperspectral method uses a broader spectral range to improve accuracy while also providing estimates of temperature, emissivity profiles, and contributions of downwelling from a collection of zenith angles. Experimental results demonstrate improved ranging accuracy, in one case reducing error from over 100 m when reflected light is not modeled to 6.8 m with the quadspectral method and 1.2 m with the hyperspectral method.

cs.CV↗

arXiv2025

Benchmarking Nighttime Traffic Sign Recognition with Illumination-Adaptive Detection and Semantic Attribute Reasoning

Traffic signboards are vital for road safety and intelligent transportation systems. Yet, recognizing traffic signs at night remains underexplored due to the scarcity of real-world public datasets capturing low-light degradations and distractor classes. Existing benchmarks are predominantly daytime and do not reflect challenges such as headlight glare, motion blur, sensor noise, and vandalized or ambiguous signage. To address these gaps, we introduce INTSD, a large-scale nighttime traffic sign dataset collected across diverse regions of India. INTSD contains street-level images spanning 41 traffic signboard classes, multiple distractor categories, and varied lighting and weather conditions, designed to support both detection and fine-grained classification under nighttime scenarios. To benchmark INTSD, we conduct extensive evaluations using state-of-the-art detection and classification models under standardized protocols. Additionally, we present LENS-Net, a strong baseline that integrates an end-to-end adaptive illumination-aware detector with a multimodal classifier that fuses vision-language representations with soft semantic attribute reasoning over learnable shape and color embeddings. Experiments demonstrate that models trained exclusively on daytime data fail substantially under real nighttime conditions - a gap that is recovered once INTSD is introduced in training, even when controlling for data volume. These results validate INTSD as a complementary nighttime training resource and establish competitive baselines for future research. The code and dataset are publicly available.

cs.CV↗

arXiv2025

Image Reconstruction from Readout-Multiplexed Single-Photon Detector Arrays

Readout multiplexing is a promising solution to overcome hardware limitations and data bottlenecks in imaging with single-photon detectors. Conventional multiplexed readout processing creates an upper bound on photon counts at a very fine time scale, where frames with multiple detected photons must either be discarded or allowed to introduce significant bias. We formulate multiphoton coincidence resolution as an inverse imaging problem and introduce a solution framework to probabilistically resolve the spatial locations of photon incidences. Specifically, we develop a theoretical abstraction of row--column multiplexing and a model of photon events that make readouts ambiguous. Using this, we propose a novel estimator that spatially resolves up to four coincident photons. Monte Carlo simulations show that our proposed method increases the peak signal-to-noise ratio (PSNR) of reconstruction by 3 to 4 dB compared to conventional methods under optimal incident flux conditions. Additionally, this method reduces the required number of readout frames to achieve the same mean-squared error as other methods by a factor of ~4. Finally, our method matches the Cramer--Rao bound for detection probability estimation for a wider range of incident flux values compared to conventional methods. While demonstrated for a specific detector type and readout architecture, this method can be extended to more general multiplexing with different detector models.

quant-ph↗

arXiv2024

Shot noise-mitigated secondary electron imaging with ion count-aided microscopy

Modern science is dependent on imaging on the nanoscale, often achieved through processes that detect secondary electrons created by a highly focused incident charged particle beam. Multiple types of measurement noise limit the ultimate trade-off between the image quality and the incident particle dose, which can preclude useful imaging of dose-sensitive samples. Existing methods to improve image quality do not fundamentally mitigate the noise sources. Furthermore, barriers to assigning a physically meaningful scale make the images qualitative. Here we introduce ion count-aided microscopy (ICAM), which is a quantitative imaging technique that uses statistically principled estimation of the secondary electron yield. With a readily implemented change in data collection, ICAM substantially reduces source shot noise. In helium ion microscopy, we demonstrate 3x dose reduction and a good match between these empirical results and theoretical performance predictions. ICAM facilitates imaging of fragile samples and may make imaging with heavier particles more attractive.

physics.med-ph↗

arXiv2024

A Novel Sector-Based Algorithm for an Optimized Star-Galaxy Classification

This paper introduces a novel sector-based methodology for star-galaxy classification, leveraging the latest Sloan Digital Sky Survey data (SDSS-DR18). By strategically segmenting the sky into sectors aligned with SDSS observational patterns and employing a dedicated convolutional neural network (CNN), we achieve state-of-the-art performance for star galaxy classification. Our preliminary results demonstrate a promising pathway for efficient and precise astronomical analysis, especially in real-time observational settings.

astro-ph.IM↗

arXiv2023

Continuous-Time Modeling and Analysis of Particle Beam Metrology

Particle beam microscopy (PBM) performs nanoscale imaging by pixelwise capture of scalar values representing noisy measurements of the response from secondary electrons (SEs) integrated over a dwell time. Extended to metrology, goals include estimating SE yield at each pixel and detecting differences in SE yield across pixels; obstacles include shot noise in the particle source as well as lack of knowledge of and variability in the instrument response to single SEs. A recently introduced time-resolved measurement paradigm promises mitigation of source shot noise, but its analysis and development have been largely limited to estimation problems under an idealization in which SE bursts are directly and perfectly counted. Here, analyses are extended to error exponents in feature detection problems and to degraded measurements that are representative of actual instrument behavior for estimation problems. For estimation from idealized SE counts, insights on existing estimators and a superior estimator are also provided. For estimation in a realistic PBM imaging scenario, extensions to the idealized model are introduced, methods for model parameter extraction are discussed, and large improvements from time-resolved data are presented.

physics.data-an↗

arXiv2021

Online Beam Current Estimation in Particle Beam Microscopy

In conventional particle beam microscopy, knowledge of the beam current is essential for accurate micrograph formation and sample milling. This generally necessitates offline calibration of the instrument. In this work, we establish that beam current can be estimated online, from the same secondary electron count data that is used to form micrographs. Our methods depend on the recently introduced time-resolved measurement concept, which combines multiple short measurements at a single pixel and has previously been shown to partially mitigate the effect of beam current variation on micrograph accuracy. We analyze the problem of jointly estimating beam current and secondary electron yield using the Cramer-Rao bound. Joint estimators operating at a single pixel and estimators that exploit models for inter-pixel correlation and Markov beam current variation are proposed and tested on synthetic microscopy data. Our estimates of secondary electron yield that incorporate explicit beam current estimation beat state-of-the-art methods, resulting in micrograph accuracy nearly indistinguishable from what is obtained with perfect beam current knowledge. Our novel beam current estimation could help improve milling outcomes, prevent sample damage, and enable online instrument diagnostics.

physics.med-ph↗

arXiv2021

Secondary Electron Count Imaging in SEM

Scanning electron microscopy (SEM) is a versatile technique used to image samples at the nanoscale. Conventional imaging by this technique relies on finding the average intensity of the signal generated on a detector by secondary electrons (SEs) emitted from the sample and is subject to noise due to variations in the voltage signal from the detector. This noise can result in degradation of the SEM image quality for a given imaging dose. SE count imaging, which uses the direct count of SEs detected from the sample instead of the average signal intensity, would overcome this limitation and lead to improvement in SEM image quality. In this paper, we implement an SE count imaging scheme by synchronously outcoupling the detector and beam scan signals from the microscope and using custom code to count detected SEs. We demonstrate a ~30% increase in the image signal-to-noise-ratio due to SE counting compared to conventional imaging. The only external hardware requirement for this imaging scheme is an oscilloscope fast enough to accurately sample the detector signal for SE counting, making the scheme easily implementable on any SEM.

physics.ins-det↗

arXiv2021

MD-CSDNetwork: Multi-Domain Cross Stitched Network for Deepfake Detection

The rapid progress in the ease of creating and spreading ultra-realistic media over social platforms calls for an urgent need to develop a generalizable deepfake detection technique. It has been observed that current deepfake generation methods leave discriminative artifacts in the frequency spectrum of fake images and videos. Inspired by this observation, in this paper, we present a novel approach, termed as MD-CSDNetwork, for combining the features in the spatial and frequency domains to mine a shared discriminative representation for classifying \textit{deepfakes}. MD-CSDNetwork is a novel cross-stitched network with two parallel branches carrying the spatial and frequency information, respectively. We hypothesize that these multi-domain input data streams can be considered as related supervisory signals. The supervision from both branches ensures better performance and generalization. Further, the concept of cross-stitch connections is utilized where they are inserted between the two branches to learn an optimal combination of domain-specific and shared representations from other domains automatically. Extensive experiments are conducted on the popular benchmark dataset namely FaceForeniscs++ for forgery classification. We report improvements over all the manipulation types in FaceForensics++ dataset and comparable results with state-of-the-art methods for cross-database evaluation on the Celeb-DF dataset and the Deepfake Detection Dataset.

cs.CV↗

arXiv2021

Evaluating Empathetic Chatbots in Customer Service Settings

Customer service is a setting that calls for empathy in live human agent responses. Recent advances have demonstrated how open-domain chatbots can be trained to demonstrate empathy when responding to live human utterances. We show that a blended skills chatbot model that responds to customer queries is more likely to resemble actual human agent response if it is trained to recognize emotion and exhibit appropriate empathy, than a model without such training. For our analysis, we leverage a Twitter customer service dataset containing several million customer<->agent dialog examples in customer service contexts from 20 well-known brands.

cs.CL↗

arXiv2020

WaveTransform: Crafting Adversarial Examples via Input Decomposition

Frequency spectrum has played a significant role in learning unique and discriminating features for object recognition. Both low and high frequency information present in images have been extracted and learnt by a host of representation learning techniques, including deep learning. Inspired by this observation, we introduce a novel class of adversarial attacks, namely `WaveTransform', that creates adversarial noise corresponding to low-frequency and high-frequency subbands, separately (or in combination). The frequency subbands are analyzed using wavelet decomposition; the subbands are corrupted and then used to construct an adversarial example. Experiments are performed using multiple databases and CNN models to establish the effectiveness of the proposed WaveTransform attack and analyze the importance of a particular frequency component. The robustness of the proposed attack is also evaluated through its transferability and resiliency against a recent adversarial defense algorithm. Experiments show that the proposed attack is effective against the defense algorithm and is also transferable across CNNs.

cs.CV↗

arXiv2020

Superconducting MoN thin films prepared by DC reactive magnetron sputtering for nanowire single-photon detectors

We present a comprehensive study of molybdenum nitride (MoN) thin film deposition using direct current (DC) reactive magnetron sputtering. We have investigated the effect of various deposition conditions on the superconducting and electrical properties of the films. Furthermore, we have shown that meander-shaped single-photon detectors made from 5 nm MoN films have saturated quantum detection efficiency at the telecom wavelength of 1550 nm. Our results indicate that MoN may be a material of interest for practical applications of low-temperature superconductors, including single-photon detectors and transition-edge sensors.

cond-mat.supr-con↗

arXiv2020

Generalized Iris Presentation Attack Detection Algorithm under Cross-Database Settings

Presentation attacks are posing major challenges to most of the biometric modalities. Iris recognition, which is considered as one of the most accurate biometric modality for person identification, has also been shown to be vulnerable to advanced presentation attacks such as 3D contact lenses and textured lens. While in the literature, several presentation attack detection (PAD) algorithms are presented; a significant limitation is the generalizability against an unseen database, unseen sensor, and different imaging environment. To address this challenge, we propose a generalized deep learning-based PAD network, MVANet, which utilizes multiple representation layers. It is inspired by the simplicity and success of hybrid algorithm or fusion of multiple detection networks. The computational complexity is an essential factor in training deep neural networks; therefore, to reduce the computational complexity while learning multiple feature representation layers, a fixed base model has been used. The performance of the proposed network is demonstrated on multiple databases such as IIITD-WVU MUIPA and IIITD-CLI databases under cross-database training-testing settings, to assess the generalizability of the proposed algorithm.

cs.CV↗

arXiv2020

MixNet for Generalized Face Presentation Attack Detection

The non-intrusive nature and high accuracy of face recognition algorithms have led to their successful deployment across multiple applications ranging from border access to mobile unlocking and digital payments. However, their vulnerability against sophisticated and cost-effective presentation attack mediums raises essential questions regarding its reliability. In the literature, several presentation attack detection algorithms are presented; however, they are still far behind from reality. The major problem with existing work is the generalizability against multiple attacks both in the seen and unseen setting. The algorithms which are useful for one kind of attack (such as print) perform unsatisfactorily for another type of attack (such as silicone masks). In this research, we have proposed a deep learning-based network termed as \textit{MixNet} to detect presentation attacks in cross-database and unseen attack settings. The proposed algorithm utilizes state-of-the-art convolutional neural network architectures and learns the feature mapping for each attack category. Experiments are performed using multiple challenging face presentation attack databases such as SMAD and Spoof In the Wild (SiW-M) databases. Extensive experiments and comparison with existing state of the art algorithms show the effectiveness of the proposed algorithm.

cs.CV↗