3D Vision & Remote Sensing Lab

We're a cutting-edge research laboratory led by Dr. Emon Kumar Dey, dedicated to advancing the frontiers of 3D Point Cloud Processing, Remote Sensing, and Computer Vision through innovative research and development. We develop next-generation AI systems that understand and interact with the three-dimensional world.

Explore Research Publications

Research Areas

Our lab focuses on three interconnected domains that define the future of spatial computing and environmental monitoring through advanced AI and computer vision techniques.

3D Point Cloud Processing

Advanced algorithms for point cloud segmentation, classification, and 3D object recognition. We develop novel deep learning architectures for processing large-scale 3D data from LiDAR and photogrammetry sources.

  • Point cloud segmentation
  • 3D object detection
  • Surface reconstruction
  • Real-time processing

Remote Sensing

Leveraging satellite imagery, aerial photography, and sensor networks for environmental monitoring, urban planning, and precision agriculture. Our work spans from data acquisition to automated analysis pipelines.

  • Multispectral analysis
  • 3D scene understanding
  • Environmental monitoring
  • Urban planning

Computer Vision & AI Systems

Developing next-generation AI and computer vision technologies for intelligent perception, transparent decision-making, and real-world applications.

  • Pattern recognition
  • Features Analysis
  • Explainable AI
  • Causal AI

Recent Publications

Our latest contributions to the scientific community in top-tier conferences and journals, pushing the boundaries of 3D vision and remote sensing research.

Roof Boundary Points Extraction From LiDAR Point Cloud Data Using Adaptive Neighbourhood-Based α-Shape Algorithm

Authors: Emon Kumar Dey | Journal: IEEE Access | Quartile: Q1 | Impact Factor: 3.6 | Year: 2025 |

Proposed a boundary point extraction technique that selects optimal α values adaptively for individual points by analyzing local geometric point density using an adaptive variable point neighbourhood selection technique.

Point Cloud α-Shape Boundary Extraction

Segmentation of LiDAR Point Cloud Data in Urban Areas Using Adaptive Neighborhood Selection Technique

Authors: Debobrata Chakraborty, Emon Kumar Dey | Journal: Plos One | Quartile: Q1 | Impact Factor: 2.6 | Year: 2024 |

Proposed an suitable adaptive neighborhood selection approach by completely considering the complex and heterogeneous nature of the input LiDAR point cloud data in urban area for point cloud segmentation, achieving state-of-the-art performance on Vaihingen and Torronto-3D dataset.

Point Cloud Neighborhood Segmentation

Research Projects

Explore our ongoing research initiatives that are pushing the boundaries of 3D vision, AI systems, and remote sensing technologies.

Ongoing Projects

Building boundary point extraction from LiDAR Point cloud data

Point Cloud Building Roof Extraction Ph-shape

The Fusion Biomarker Network: An Adversarial Model for Early Alzheimer’s Disease Prediction using Multi Modal Neuroimaging

Medical Imaging Adversarial

Low-Light Image Enhancement by Multi-Scale Fusion Technique

Image Processing Low-light

Completed Projects

Research Grants

Our research is supported by prestigious funding agencies and organizations that recognize the impact of our work in advancing 3D vision and AI technologies.

Effective Building Roof Boundary Extraction from Segmented LiDAR Point Cloud Data

Funding Agency: ICT Innovation fund| Year: 2023-24 |

Persistent Homology-based Adaptive Approach for Building Roof Outline Extraction from Airborne LiDAR Point Cloud Data

Funding Agency: UGC Research Grant | Year: 2024-25 |

Explainable Stacked Ensembles Learning for LiDAR Point Cloud Segmentation

Funding Agency: ICT Innovation Fund | Year: 2024-25 |

Airborne 3D LiDAR Point Cloud Classification By Using Machine Learning

Funding Agency: ICT Innovation Fund | Year: 2023-24|

Our Team

Meet the brilliant minds driving innovation in 3D vision and remote sensing research. Our diverse team combines expertise in AI, computer vision, and geospatial sciences.

Dr. Emon Kumar Dey

Associate Professor at University of Dhaka

Principal Investigator

Ph.D. from Griffith University, Australia (2022) specializing in automated building extraction and evaluation using LiDAR point clouds data, image, or fusion. Expert in object classification, remote sensing analysis, and machine learning applications for geospatial intelligence.

Research Group Member

Debobrata Chakraborty

Currently: Lecturer at City University

Master's Student (MSSE-10)

Afridi Rahman Bondhon

Master's Student (MSSE-11)

Ahmed Adnan

Currently: Lecturer at East West University

Master's Student (MSSE-11)

Mushfiqur Rahman Chowdhury

Currently: Lecturer at Daffodil International University

Undergrad Student (BSSE-11)

Soumitra Paul

Undergrad Student (BSSE-13)

Md. Rejaul karim

Undergrad Student (BSSE-13)

Connect With Us

Interested in collaboration, joining our research team, or have questions about our work? We'd love to hear from you and explore potential partnerships.

Address

Institute of Information Technology
University of Dhaka
Suhrawardi Udyan Rd, Dhaka 1200