arXiv subjects

Himanshu Kumar

Publications and source records attributed to Himanshu Kumar.

29 records · Page 2Linked to original sources

arXiv2025

3rd Workshop on Maritime Computer Vision (MaCVi) 2025: Challenge Results

The 3rd Workshop on Maritime Computer Vision (MaCVi) 2025 addresses maritime computer vision for Unmanned Surface Vehicles (USV) and underwater. This report offers a comprehensive overview of the findings from the challenges. We provide both statistical and qualitative analyses, evaluating trends from over 700 submissions. All datasets, evaluation code, and the leaderboard are available to the public at https://macvi.org/workshop/macvi25.

cs.CV↗

arXiv2025

A Multimodal Dataset for Enhancing Industrial Task Monitoring and Engagement Prediction

Detecting and interpreting operator actions, engagement, and object interactions in dynamic industrial workflows remains a significant challenge in human-robot collaboration research, especially within complex, real-world environments. Traditional unimodal methods often fall short of capturing the intricacies of these unstructured industrial settings. To address this gap, we present a novel Multimodal Industrial Activity Monitoring (MIAM) dataset that captures realistic assembly and disassembly tasks, facilitating the evaluation of key meta-tasks such as action localization, object interaction, and engagement prediction. The dataset comprises multi-view RGB, depth, and Inertial Measurement Unit (IMU) data collected from 22 sessions, amounting to 290 minutes of untrimmed video, annotated in detail for task performance and operator behavior. Its distinctiveness lies in the integration of multiple data modalities and its emphasis on real-world, untrimmed industrial workflows-key for advancing research in human-robot collaboration and operator monitoring. Additionally, we propose a multimodal network that fuses RGB frames, IMU data, and skeleton sequences to predict engagement levels during industrial tasks. Our approach improves the accuracy of recognizing engagement states, providing a robust solution for monitoring operator performance in dynamic industrial environments. The dataset and code can be accessed from https://github.com/navalkishoremehta95/MIAM/.

cs.CV↗

arXiv2024

Quantum gravitational corrections to the geometry of charged AdS black holes

We study the quantum gravitational corrections to the geometry of a four-dimensionalcharged (Reissner-Nordström) Anti de Sitter black hole starting from an effective field theory approach to quantum gravity. We use the expression of the modified horizon radius to compute the quantum corrected Wald entropy, whose expression reproduces the logarithmic behaviour found by other methods. We perform a thermodynamics analysis and compute the quantum gravitational corrections to the temperature, pressure, specific heat and Helmholtz free energy. All these quantities are renormalisation group invariant. We find that a quantum charged AdS black hole can exist only for a bounded range of masses and that it can undergo a second order phase transition as it moves from a state with positive specific heat to a negative one.

hep-th↗

arXiv2024

High dynamic-range and portable magnetometer using ensemble nitrogen-vacancy centers in diamond

Nitrogen vacancy (NV) centers in diamonds have been explored for realizing a wide range of sensing applications in the last decade due to their unique quantum properties. Here we realize a compact and portable magnetometer with an ensemble of NV centers which we call the Quantum MagPI (Quantum Magnetometer with Proportional Integral control). Including the sensor head and associated electronics, our sensor assembly can fit inside 10 cm x 10 cm x 7 cm box and control electronics in 30 cm x 25 cm x 5 cm box. We achieve a bandwidth normalized sensitivity of ~ 10 nT/sqrt(Hz). Using closed-loop feedback for locking to the resonance frequency, we extend the linear dynamic range to 200 microT (20x improvement compared to the intrinsic dynamic range) without compromising the sensitivity. We report a detailed performance analysis of the magnetometer through measurements of noise spectra, Allan deviation, and tracking of nT-level magnetic fields in real-time. Additionally, we demonstrate the utility of such a magnetometer by real-time tracking the movement of the elevator car and door opening by measuring the projection of the magnetic field along one of the NV-axes under ambient temperature and humidity.

quant-ph↗

arXiv2024

A Volumetric Saliency Guided Image Summarization for RGB-D Indoor Scene Classification

Image summary, an abridged version of the original visual content, can be used to represent the scene. Thus, tasks such as scene classification, identification, indexing, etc., can be performed efficiently using the unique summary. Saliency is the most commonly used technique for generating the relevant image summary. However, the definition of saliency is subjective in nature and depends upon the application. Existing saliency detection methods using RGB-D data mainly focus on color, texture, and depth features. Consequently, the generated summary contains either foreground objects or non-stationary objects. However, applications such as scene identification require stationary characteristics of the scene, unlike state-of-the-art methods. This paper proposes a novel volumetric saliency-guided framework for indoor scene classification. The results highlight the efficacy of the proposed method.

cs.CV↗

arXiv2022

ParaColorizer: Realistic Image Colorization using Parallel Generative Networks

Grayscale image colorization is a fascinating application of AI for information restoration. The inherently ill-posed nature of the problem makes it even more challenging since the outputs could be multi-modal. The learning-based methods currently in use produce acceptable results for straightforward cases but usually fail to restore the contextual information in the absence of clear figure-ground separation. Also, the images suffer from color bleeding and desaturated backgrounds since a single model trained on full image features is insufficient for learning the diverse data modes. To address these issues, we present a parallel GAN-based colorization framework. In our approach, each separately tailored GAN pipeline colorizes the foreground (using object-level features) or the background (using full-image features). The foreground pipeline employs a Residual-UNet with self-attention as its generator trained using the full-image features and the corresponding object-level features from the COCO dataset. The background pipeline relies on full-image features and additional training examples from the Places dataset. We design a DenseFuse-based fusion network to obtain the final colorized image by feature-based fusion of the parallelly generated outputs. We show the shortcomings of the non-perceptual evaluation metrics commonly used to assess multi-modal problems like image colorization and perform extensive performance evaluation of our framework using multiple perceptual metrics. Our approach outperforms most of the existing learning-based methods and produces results comparable to the state-of-the-art. Further, we performed a runtime analysis and obtained an average inference time of 24ms per image.

cs.CV↗

arXiv2021

DSRN: an Efficient Deep Network for Image Relighting

Custom and natural lighting conditions can be emulated in images of the scene during post-editing. Extraordinary capabilities of the deep learning framework can be utilized for such purpose. Deep image relighting allows automatic photo enhancement by illumination-specific retouching. Most of the state-of-the-art methods for relighting are run-time intensive and memory inefficient. In this paper, we propose an efficient, real-time framework Deep Stacked Relighting Network (DSRN) for image relighting by utilizing the aggregated features from input image at different scales. Our model is very lightweight with total size of about 42 MB and has an average inference time of about 0.0116s for image of resolution $1024 \times 1024$ which is faster as compared to other multi-scale models. Our solution is quite robust for translating image color temperature from input image to target image and also performs moderately for light gradient generation with respect to the target image. Additionally, we show that if images illuminated from opposite directions are used as input, the qualitative results improve over using a single input image.

cs.CV↗

arXiv2021

Fast and Accurate Quantized Camera Scene Detection on Smartphones, Mobile AI 2021 Challenge: Report

Camera scene detection is among the most popular computer vision problem on smartphones. While many custom solutions were developed for this task by phone vendors, none of the designed models were available publicly up until now. To address this problem, we introduce the first Mobile AI challenge, where the target is to develop quantized deep learning-based camera scene classification solutions that can demonstrate a real-time performance on smartphones and IoT platforms. For this, the participants were provided with a large-scale CamSDD dataset consisting of more than 11K images belonging to the 30 most important scene categories. The runtime of all models was evaluated on the popular Apple Bionic A11 platform that can be found in many iOS devices. The proposed solutions are fully compatible with all major mobile AI accelerators and can demonstrate more than 100-200 FPS on the majority of recent smartphone platforms while achieving a top-3 accuracy of more than 98%. A detailed description of all models developed in the challenge is provided in this paper.

eess.IV↗

arXiv2018

Investigation of Ta2O5 as an alternative high \k{appa} dielectric for InAlN/GaN MOS HEMT on Si

We report on the demonstration and investigation of Ta2O5 as high-\k{appa} dielectric for InAlN/GaN-MOS HEMT-on-Si. Ta2O5 of thickness 24 nm and dielectric constant ~ 30 was sputter deposited on InAlN/GaN HEMT and was investigated for different post deposition anneal conditions (PDA). The gate leakage was 16nA/mm at -15 V which was ~ 5 orders of magnitude lower compared to reference HEMT. The 2-dimensional electron gas (2DEG) density was found to vary with annealing temperature suggesting the presence of net charge at the Ta2O5/InAlN interface. Dispersion in the capacitance-voltage (C-V) characteristics was used to estimate the frequency-dependent interface charge while energy band diagrams under flat band conditions were investigated to estimate fixed charge. The optimum anneal condition was found to be 500° C which has resulted into a flat band voltage spread (VFB) of 0.4 V and interface fix charge (Qf) of 3.98x10^13 cm-2. XPS (X-ray photoelectron spectroscopy) spectra of as deposited and annealed Ta2O5 film were analyzed for Ta and O compositions in the film. The sample annealed at 500° C has shown Ta:O ratio of 0.41.XRD (X-ray diffraction) analysis was done to check the evolution of poly-crystallization of the Ta2O5 film at higher annealing temperatures.

physics.app-ph↗

arXiv2017

Robust Loss Functions under Label Noise for Deep Neural Networks

In many applications of classifier learning, training data suffers from label noise. Deep networks are learned using huge training data where the problem of noisy labels is particularly relevant. The current techniques proposed for learning deep networks under label noise focus on modifying the network architecture and on algorithms for estimating true labels from noisy labels. An alternate approach would be to look for loss functions that are inherently noise-tolerant. For binary classification there exist theoretical results on loss functions that are robust to label noise. In this paper, we provide some sufficient conditions on a loss function so that risk minimization under that loss function would be inherently tolerant to label noise for multiclass classification problems. These results generalize the existing results on noise-tolerant loss functions for binary classification. We study some of the widely used loss functions in deep networks and show that the loss function based on mean absolute value of error is inherently robust to label noise. Thus standard back propagation is enough to learn the true classifier even under label noise. Through experiments, we illustrate the robustness of risk minimization with such loss functions for learning neural networks.

stat.ML↗

arXiv2013

Feebly Self-Interacting Cold Dark Matter: New theory for the Core-Halo structure in GLSB Galaxies

We explore the low energy cosmological dynamics of feebly self-interacting cold dark matter and propose a new simple explanation for the rotation curves of the core-halo model in massive LSB (Low Surface brightness)galaxies. We argue in favor of the truly collisionless nature of cold dark matter,which is feebly,self-interacting at small scales between epochs of equality and recombination.For this, we assume a model, wherein strongly coupled baryon-radiation plasma ejects out of small regions of concentrated cold dark matter without losing its equilibrium. We use the Merscerskii equation i.e. the variable mass formalism of classical dynamics.We obtain new results relating the oscillations in the CMB anisotropy to the ejection velocity of the baryon-radiation plasma,which can be useful tool for numerical work for exploring the second peak of CMB. Based on this model, we discuss the growth of perturbations in such a feebly self-interacting,cold dark matter both in the Jeans theory and in the expanding universe using Newton's theory.We obtain an expression for the growth of fractional perturbations in cold dark matter,which reduce to the standard result of perturbation theory for late recombination epochs. We see the effect of the average of the perturbations in the cold dark matter potential on the cosmic microwave background temperature anisotropy that originated at redshifts between equality and recombination i.e. 1100 < z < z_{eq}. Also we obtain an expression for the Sachs-Wolfe effect,i.e. the CMB temperature anisotropy at decoupling in terms of the average of the perturbations in cold dark matter potential.

astro-ph.CO↗