A5011222429- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- RNA Research and Splicing
- Advanced Vision and Imaging
- Chaos-based Image/Signal Encryption
- Robotic Path Planning Algorithms
- Fractal and DNA sequence analysis
- RNA and protein synthesis mechanisms
- RNA modifications and cancer
- Vehicle License Plate Recognition
- Dental Radiography and Imaging
- Cryptographic Implementations and Security
- Molecular Communication and Nanonetworks
- Infrastructure Maintenance and Monitoring
- MicroRNA in disease regulation
- Endodontics and Root Canal Treatments
- Neuroinflammation and Neurodegeneration Mechanisms
- Opportunistic and Delay-Tolerant Networks
- Single-cell and spatial transcriptomics
- Dental materials and restorations
- Underwater Vehicles and Communication Systems
- Currency Recognition and Detection
- Cell Image Analysis Techniques
University at Buffalo, State University of New York
2024
People’s University
2024
University of Luxembourg
2020-2023
National Institute of Technology Agartala
2023
Indian Institute of Science Education and Research, Bhopal
2021
International Centre for Genetic Engineering and Biotechnology
2018
Off-road image semantic segmentation is challenging due to the presence of uneven terrain, unstructured class boundaries, irregular features and strong textures. These aspects affect vehicle perception. Current off-road datasets exhibit difficulties like imbalance understanding varying environmental topography. To overcome these issues, we propose a framework for (OFFSEG) that involves (i) pooled with four classes (sky, traversable region, non-traversable region obstacle) using...
Summary Previously, we reported that infection of human macrophages with Mycobacterium tuberculosis ( Mtb ) results in massive alterations the pattern RNA splicing host. The finding gained significance since alternate spliced variants a same gene may have substantially different structure, function, stability, interaction partners, localization, and so forth, owing to inclusion or exclusion specific exons. To establish proof‐of‐concept; on how infection‐induced could impact protein...
The functional specialization of cell types arises during development and is shaped by cell-cell communication networks determining a distribution states that are collectively important for tissue functioning. However, the identification these tissue-specific remains challenging. Although plethora computational approaches have been successful in detecting subtypes, they fail resolving states. To address this issue, we present FunRes, method designed FunRes relies on scRNA-seq data to...
We present OffRoadTranSeg, the first end-to-end framework for semi-supervised segmentation in unstructured outdoor environment using transformers and automatic data selection labelling. The offroad is a scene understanding approach that widely used autonomous driving. popular method to use fully connected convolution layers large labelled data, however, due class imbalance, there will be several mismatches also some classes may not detected. Our do task of manner. aim provide model where...
Astrocytes arise from multipotent neural stem cells (NSCs) and represent the most abundant cell type of central nervous system (CNS), playing key roles in developing adult brain. Since differentiation NSCs towards a gliogenic fate is precisely timed regulated process, its perturbation gives rise to dysfunctional astrocytic phenotypes. Inflammation, which often underlies neurological disorders, including neurodevelopmental disorders brain tumors, disrupts accurate developmental process NSCs....
Nature has evolved humans to walk on different terrains by developing a detailed understanding of their physical characteristics. Similarly, legged robots need develop capability complex with variety task-dependent payloads achieve goals. However, conventional terrain adaptation methods are susceptible failure varying payloads. In this work, we introduce PANOS, weakly supervised approach that integrates proprioception and exteroception from onboard sensing stable gait while walking robot...
Historically, feature-based approaches have been used extensively for camera-based robot perception tasks such as localization, mapping, tracking, and others. Several of these also combine other sensors (inertial sensing, example) to perform combined state estimation. Our work rethinks this approach; we present a representation learning mechanism that identifies visual features best correspond motion estimated by an external signal. Specifically, utilize the robot's transformations through...