A5101630690- Robotic Locomotion and Control
- Gait Recognition and Analysis
- Mobile Ad Hoc Networks
- Human Pose and Action Recognition
- Data Mining Algorithms and Applications
- Muscle activation and electromyography studies
- Vehicular Ad Hoc Networks (VANETs)
- Opportunistic and Delay-Tolerant Networks
- Advanced Image and Video Retrieval Techniques
- Magnetic and transport properties of perovskites and related materials
- Robotic Path Planning Algorithms
- Advanced Neural Network Applications
- Domain Adaptation and Few-Shot Learning
- Robotic Mechanisms and Dynamics
- Data Management and Algorithms
- Robot Manipulation and Learning
- Energy Efficient Wireless Sensor Networks
- Wireless Networks and Protocols
- Industrial Vision Systems and Defect Detection
- Generative Adversarial Networks and Image Synthesis
- Advanced Text Analysis Techniques
- Context-Aware Activity Recognition Systems
- Prosthetics and Rehabilitation Robotics
- Multimodal Machine Learning Applications
- EEG and Brain-Computer Interfaces
Malaviya National Institute of Technology Jaipur
2021-2024
GLA University
2022-2024
LNM Institute of Information Technology
2023
Bharati Vidyapeeth Deemed University
2023
Fort Valley State University
1999-2017
Indian Institute of Information Technology Allahabad
2015
Bir Hospital
2015
Jawaharlal Nehru University
2014
Petronas (Malaysia)
2014
Barkatullah University
2010-2013
With the growing importance of large network models and enormous training datasets, GPUs have become increasingly necessary to train neural networks. This is largely because conventional optimization algorithms rely on stochastic gradient methods that don't scale well numbers cores in a cluster setting. Furthermore, convergence all methods, including batch suffers from common problems like saturation effects, poor conditioning, saddle points. paper explores an unconventional method uses...
ABSTRACT Purpose Pharmacovigilance methods have advanced greatly during the last decades, making post‐market drug assessment an essential evaluation component. These mainly rely on use of spontaneous reporting systems and health information databases to collect expertise from huge amounts real‐world reports. The EU‐ADR Web Platform was built further facilitate accessing, monitoring exploring these data, enabling in‐depth analysis adverse reactions risks. Methods exploits wealth data...
Large-scale pre-training tasks like image classification, captioning, or self-supervised techniques do not incentivize learning the semantic boundaries of objects. However, recent generative foundation models built using text-based latent diffusion may learn boundaries. This is because they have to synthesize intricate details about all objects in an based on a text description. Therefore, we present technique for segmenting real and AI-generated images (LDMs) trained internet-scale...
Visible Light Communication (VLC) has emerged as a promising technology for high-speed wireless communication, which employs the visible spectrum to transmit information. As title indicates, IEEE 802.15.7 is new standard dedicated VLC networks that introduces specializations match characteristics and needs of this medium. In networks, one key challenges scheduling Guaranteed Time Slots (GTS) efficiently better network performance minimum end-to-end delay while delivering reliable data. The...
Lower limb activity recognition utilizing body sensor data has attracted researchers due to its practical applications, such as neuromuscular disease detection and kinesiological investigations. The employment of wearable sensors including accelerometers, gyroscopes, surface electromyography grown their low cost broad applicability. Electromyography (EMG) are preferable for automated control a lower exoskeleton or prosthesis since they detect the signal beforehand allow faster movement...
Non-maximum suppression is an integral part of the object detection pipeline. First, it sorts all boxes on basis their scores. The box M with maximum score selected and other a significant overlap (using pre-defined threshold) are suppressed. This process recursively applied remaining boxes. As per design algorithm, if lies within predefined threshold, leads to miss. To this end, we propose Soft-NMS, algorithm which decays scores objects as continuous function M. Hence, no eliminated in...
Monitoring temperature parameters in Permanent Magnet Surface Machine (PMSM) is crucial since the machine possesses applications several diverse areas, for instance, traction drives, electric vehicles, besides that require control of rotor order to ensure safety and cost-effectiveness. The thermal losses produced permanent magnet synchronous such as copper, iron, mechan-ical loss cooling modes are responsible rise PMSM. basic method evaluate temper-ature internal components like Lumped...
In this paper proposes a Randomized Virtual Scanning Technique to optimize the performance for road network surveillance.The Sensor it may be scan area of network.The is show as graph.So through genetic algorithm.The main focus improve energy and sensor scanning process.The sensing programming rule plays job target intrusion detection, utilizing distinctive options networks.Sensor alternate sequence node so that's why 100% detection guaranty w seconds (i.e single takes time area) along...
Robotic manipulators have become the key component in automation industry because of their accuracy. Finding inverse kinematic (IK) solution (to compute joint angles for desired end-effector position and orientation) a robot manipulator is most challenging problem. Conventional approaches like numerical, geometric, algebraic failed to provide due redundancy/singularity present IK. This paper presents IK 5-DoFs manipu-lator using fourteen meta-heuristic techniques. The determined by solving...
VANETs provide the ability for vehicles to spontaneously and wirelessly network with other nearby purposes of providing travelers new features applications. Dense Traffic has become major issue at rush-hours traffic in big cities where congested roads cost over billions lost worker productivity gallons fuel due incidents. We dispense prosper a framework early detection incident dense using vehicular ad-hoc networks. proposed techniques Incident Detections Node (IDN) collect On-Board Unit...
Translation is one of the needs global society for communicating thoughts and ideas country with other country. process interpretation text meaning subsequent production equivalent text, also called as same (message) in another language. In this paper we gave detail information on how to convert source language target using Transfer Based Approach machine translation. Here implemented English Sanskrit translator transfer based approach. used business communication but large amount population...