Embedded Files
Contact:
Contact:
3D GAIL RS Lab (3D Geospatial AI and LiDAR Remote Sensing lab)
Indian Institute of Space Science and Technology
Valiamala P. O., Thiruvananthapuram
Email: ramiya@iist.ac.in
About the data
About the data
The Thiruvananthapuram Aerial LiDAR Dataset (TALD) is a comprehensive benchmark dataset that provides high-resolution point cloud data from Airborne Laser Scanning (ALS) for urban planning, environmental monitoring, and land cover mapping. With an average density of 12 points per square meter, TALD covers a 9-square-kilometer area in Thiruvananthapuram, Kerala, India. The dataset is annotated into four classes: ground, shrub, tree, and buildings, capturing detailed 3D geometrical information about the Earth's surface, vegetation, and infrastructure. TALD was created as a benchmark for point cloud classification, enabling researchers to test and compare classification algorithms and facilitating advancements in 3D urban mapping. Its integration of optical imagery enhances its utility for detailed analysis and decision-making in urban environments.
Download
Download
The dataset can be accessed using the following website
https://catalogue.geospatial.org.in/dataset/021a3f5f-a7ce-4f3b-a532-bc1a2486b005
(The data is available only to Indian Research community)
To access the dataset contact :
Indian Institute of Space Science and Technology
Valiamala P. O., Thiruvananthapuram
Email: ramiya@iist.ac.in
Readme files / Instructions
Researchers and practitioners are invited to leverage TALD's expansive reservoir of data for diverse analytical pursuits. Available in versatile formats encompassing .laz, .txt, and .csv, this dataset lends itself seamlessly to myriad applications ranging from point cloud classification to 3D visualization and rendering. The Thiruvananthapuram Aerial LiDAR Dataset (TALD) offers high-resolution point cloud data collected over a 9-square-kilometer area in Thiruvananthapuram, Kerala, India. It encompasses spatial coordinates (X, Y, Z), RGB color information, Return Number, Number of Returns, Scan Angle Rank and Class label for each point, facilitating detailed 3D visualization and analysis. TALD is partitioned into nine blocks, each covering a 1km by 1km area, enabling manageable data handling and analysis. Additionally, each point is labeled with a corresponding class, distinguishing terrain into categories such as ground, shrub, trees, and houses. This classification scheme enhances the dataset's utility for various applications, including urban planning, environmental monitoring, and infrastructure development.
Acknowledgement
We acknowledge the use of the Airborne LiDAR Scanned dataset, whose acquisition was funded by Department of Science and Technology, Government of India, under the National Geospatial Programme (DST-NGP) on Spatial Urban Governance.
Team members in TALD data development
A.M. Ramiya, Jayati Vijaywargiya, , Mohana M, Ajay Kumar, Luxhna Ravi, Vishakh Adolph
Citation
J. Vijaywargiya and A. M. Ramiya, "Thiruvananthapuram Aerial LiDAR Dataset (TALD): A Benchmark for Complex Urban Point Cloud," in IEEE Access, vol. 13, pp. 42350-42363, 2025, doi: 10.1109/ACCESS.2025.3546628.
link: https://ieeexplore.ieee.org/abstract/document/10908393
Page updated
Google Sites
Report abuse