A5024268264- Remote-Sensing Image Classification
- Flood Risk Assessment and Management
- Remote Sensing in Agriculture
- Advanced Vision and Imaging
- Parallel Computing and Optimization Techniques
- Land Use and Ecosystem Services
- Remote Sensing and Land Use
- Urban Heat Island Mitigation
- Groundwater and Watershed Analysis
- Hydrology and Watershed Management Studies
- Remote Sensing and LiDAR Applications
- Advanced Image Processing Techniques
- Image Processing Techniques and Applications
- Game Theory and Applications
- Soil Geostatistics and Mapping
- Geochemistry and Geologic Mapping
- Cryospheric studies and observations
- Spectroscopy and Chemometric Analyses
- Functional Brain Connectivity Studies
- Landslides and related hazards
- Auction Theory and Applications
- Soil Moisture and Remote Sensing
- Automated Road and Building Extraction
- Advanced Image and Video Retrieval Techniques
- Smart Agriculture and AI
Indian Institute of Technology Roorkee
2016-2025
Indian Institute of Technology Delhi
2005-2024
Maharishi Markandeshwar University, Mullana
2024
Alabama College of Osteopathic Medicine
2024
National Institute of Hydrology
2022-2023
Indian Institute of Technology Ropar
2022-2023
Fairleigh Dickinson University
2022-2023
Induk University
2023
SMART Reading
2023
Savitribai Phule Pune University
2023
This paper gives an overview of the BlueGene/L Supercomputer. is a jointly funded research partnership between IBM and Lawrence Livermore National Laboratory as part United States Department Energy ASCI Advanced Architecture Research Program. Application performance scaling studies have recently been initiated with partners at number academic government institutions, including San Diego Supercomputer Center California Institute Technology. massively parallel system 65,536 nodes based on new...
When a scene is photographed many times by different people, the viewpoints often cluster along certain paths. These paths are largely specific to being photographed, and follow interesting regions viewpoints. We seek discover range of such turn them into controls for image-based rendering. Our approach takes as input large set community or personal photos, reconstructs camera viewpoints, automatically computes orbits, panoramas, canonical views, optimal between views. The can then be...
Given the scale of massively parallel systems, occurrence faults is no longer an exception but a regular event. Periodic checkpointing becoming increasingly important in these systems. However, huge memory footprints applications place severe limitations on scalability normal techniques. Incremental well researched technique that addresses concerns, most implementations require paging support from hardware and underlying operating system, which may not be always available. In this paper, we...
We present an algorithm for finding s-sparse vector x that minimizes the square-error ∥y -- Φx∥2 where Φ satisfies restricted isometry property (RIP), with isometric constant δ2s < 1/3. Our algorithm, called GraDeS (Gradient Descent Sparsification) iteratively updates as: [EQUATION]
We address the problem of developing discriminative, yet invariant, features for texture classification. Texture variations due to changes in scale are amongst hardest handle. One most successful methods dealing with such is based on choosing interest points and selecting their characteristic scales [Lazebnik et al. PAMI 2005]. However, a can be unstable many textures. Furthermore, reliance an point detector inability evaluate densely serious limitations. Fractals present mathematically well...
Deep learning techniques have enabled rapid progress in monocular depth estimation, but their quality is limited by the ill-posed nature of problem and scarcity high datasets. We estimate from a single cam-era leveraging dual-pixel auto-focus hardware that increasingly common on modern camera sensors. Classic stereo algorithms prior learning-based estimation underperform when applied this data, former due to too-strong assumptions about RGB image matching, latter not understanding optics...
Abstract. The Upper Ganga River basin is socioeconomically the most important river in India and highly stressed terms of water resources due to uncontrolled land use cover (LULC) activities. This study presents a comprehensive set analyses evaluate population growth, LULC transformations, quality nexus for sustainable development this basin. was conducted at two spatial scales: scale district scale. First, data were analyzed statistically demographic changes, followed by change detection...
Shallow depth-of-field is commonly used by photographers to isolate a subject from distracting background. However, standard cell phone cameras cannot produce such images optically, as their short focal lengths and small apertures capture nearly all-in-focus images. We present system computationally synthesize shallow with single mobile camera button press. If the image of person, we use person segmentation network separate accessories available, also dense dual-pixel auto-focus hardware,...
This cross-sectional research aims to study the effect of yoga practice on illness perception, and wellbeing healthy adults during 4-10 weeks lockdown due COVID19 outbreak. A total 668 (64.7% males, M = 28.12 years, SD 9.09 years) participated in online survey. The participants were grouped as; practitioners, other spiritual non-practitioners based their responses daily practices that they follow. Yoga practitioners further examined duration long-term, mid-term beginners. Multivariate...
The Urban Heat Island (UHI) effect is a significant concern in today’s rapidly urbanising cities, with exacerbating heatwaves’ impact, urban livelihood, and environmental well-being. This study aims to assess the cooling of blue-green spaces Bhubaneswar, India, explore their implications for mitigating UHI effects. Satellite images were processed Google Earth Engine (GEE) produce information on spaces’ land surface temperatures (LST). Normalised Difference Vegetation Index (NDVI) Modified...
This paper gives an overview of the BlueGene/L Supercomputer. is a jointly funded research partnership between IBM and Lawrence Livermore National Laboratory as part United States Department Energy ASCI Advanced Architecture Research Program. Application performance scaling studies have recently been initiated with partners at number academic government institutions,including San Diego Supercomputer Center California Institute Technology. massively parallel system 65,536 nodes based on new...
The Brahmaputra River is one of the largest alluvial rivers in world characterized by frequent bank erosion leading to channel pattern changes and shifting line. This study aimed at quantifying actual erosion/deposi- tion along within India for a period eighteen years (1990-2008). entire course Brah- maputra Assam from upstream Dibrugarh up town Dhubri near Bangladesh border stretch around 620 kms has been studied using an integrated approach Remote Sensing Geographical Information Sys- tem...
We present a novel method to train machine learning algorithms estimate scene depths from single image, by using the information provided camera's aperture as supervision. Prior works use depth sensor's outputs or images of same alternate viewpoints supervision, while our instead uses viewpoint taken with varying camera aperture. To enable effects we introduce two differentiable rendering functions that input image and predicted simulate depth-of-field caused real apertures. monocular...
The waterscape of the Upper Ganga river basin has reconfigured over last few decades due to rapid population growth, urbanization, and industrialization. It resulted in altered regional hydrology water quality basin. In this study, Soil Water Assessment Tool (SWAT) hydrological model was used investigate effects spatial urbanization on key components study region. Time series Land Use Cover (LULC) maps 1980, 2001, 2012 were prepared using object-based image analysis (OBIA) approach. To...
In Geographical Information Systems issues of scale are an increasing interest in storing health data and using these policy support. National international policies on treating HIV (Human Immunodeficiency Virus) positive women India based case counts at Voluntary Counseling Testing Centers (VCTCs). this study, carried out the Indian state Andhra Pradesh, centers located subdistricts called mandals, serving for both registration facility policies. This study hypothesizes that people may move...
We present an approach for generating face animations from large image collections of the same person. Such collections, which we call photobios, sample appearance a person over changes in pose, facial expression, hairstyle, age, and other variations. By optimizing order images are displayed cross-dissolving between them, control motion through space create compelling (e.g., render smooth transition frowning to smiling). Used this context, cross dissolve produces very strong effect; key...