07-15, 14:00–18:00 (Europe/Sarajevo), PA01
Dive into the fascinating world of LiDAR data as we transform raw point clouds into striking 3D visualizations using a suite of open-source tools. This workshop begins by exploring the ASPRS standard, helping you understand essential concepts like point classifications, return numbers, and other fundamental attributes that define LiDAR data. You’ll learn how to correctly interpret these attributes, assess data quality, and leverage them for more insightful analysis.
Moving beyond the basics, we’ll dig into powerful open-source workflows with PDAL, GRASS GIS, WhiteboxTools for data reading, manipulation and spatial processing, and with tools like QGIS, Potree and pybabylonjs for dynamic 3D rendering. We’ll also introduce TileDB as a robust storage option to manage point-clouds as massive dataframes. Along the way, we’ll tackle real-world tasks like gridding, interpolation, vectorization, and classification, highlighting effective approaches to manage and visualize large-scale point clouds in a reproducible manner.
By integrating Python scripts with these specialized tools, you’ll discover how to automate complex processing chains and generate stunning outputs that bring your LiDAR data to life. While basic Python knowledge is helpful, it’s not mandatory — anyone eager to learn new techniques can follow along. By the end of this hands-on workshop, you’ll not only grasp the core concepts behind LiDAR data but also have the practical skills to handle and present it in exciting, visually compelling ways.
The workshop will cover LiDAR data standards and attributes, data import, advanced processing, and visualization workflows. Led by experts with extensive experience in LiDAR data, including its application in large-scale production tasks, the session will explore various open-source tools, each offering unique advantages for these workflows. The diversity of tools and techniques is a standout feature of this workshop, providing flexibility and tailored solutions for different projects.
The workshop will provide a comprehensive understanding of data standards and structure, focusing on the ASPRS standard, key LiDAR attributes, and common file formats such as COPC. Participants will learn how to read and preprocess LiDAR data using Python, with a focus on transforming it into formats suitable for applications like modeling and visualization. The session will cover interpolation methods to create Digital Terrain Models (DTMs), Digital Surface Models (DSMs), and Triangulated Irregular Networks (TINs), as well as deriving height-above-ground metrics critical for terrain analysis. Participants will also explore workflows for feature extraction, including identifying buildings, vegetation, or other objects, along with techniques for gridding, semantical vectorization, and data transformation for tasks such as machine learning modeling. By the end of the workshop, participants will understand how to visualize and manage large-scale point clouds effectively, gaining the skills to develop reproducible workflows tailored to their specific needs.
Level of the workshop –intermediate
Pre-requirements for attendees –Basic GIS knowledge. Please bring a laptop with permissions to boot from a USB stick. We’ll provide an ISO (and USBs) that contain all the required software and data (a modified OSGeoLive) — no installations on your existing system needed and no need to rely on internet. If you prefer not to boot from external media, set up a virtualization environment (e.g., VirtualBox or QEMU) beforehand to run the ISO. This setup ensures a hassle-free, hands-on experience right from the start.
Coding knowledge required? –While basic Python knowledge is helpful, it’s not mandatory — anyone eager to learn new techniques can follow along. You won’t code from scratch—sample code is provided—but you should be comfortable following the process. We’ll run everything in a simple IDE like Jupyter.
Link to software source code repository –
I work at the Geodetic Institute of Slovenia in Ljubljana, contributing to various projects as a Remote Sensing/Data Analyst, GIS coordinator, and specialist. Primarily, my work revolves around the analysis of (VHR) aerial imagery, satellite imagery and LiDAR point clouds. I heavily rely on Python, GRASS GIS, QGIS, PDAL, PostgreSQL, for data torturing and distribution.
I am also a PhD candidate in the Interdisciplinary Doctoral Programme Environmental Protection at the University of Ljubljana. My research focuses on developing a synergy between hyperspectral remote sensing, machine learning, and chemical analysis for soil contamination detection and monitoring.
I’ve been building GIS solutions at the Geodetic Institute of Slovenia for over 15 years, working across the stack on everything from web mapping applications to data processing pipelines. My background is in biomedical engineering, but I found my way into geospatial tech through the field of automation — and I’ve been streamlining processes and visualizing data ever since.
I’m still more of an open source power user than a contributor — but I hope someday I’ll carve out the time to give something back to the community.