3D Object Detection Engine — Course Syllabus

Reference syllabus for the 3D Object Detection Engine course delivered by the 3D Geodata Academy. It defines the learning objectives, audience, technical requirements, the module-by-module program, the assessment scheme, the results indicators and the open-source legal framework.

"44 lessons to build a complete 3D object detection engine, from LiDAR sensing fundamentals to SAM-3D and full Python apps."

1. Course Overview

Dimension	Details
Format	Self-paced online course delivered through the 3D Geodata Academy LMS.
Price	€197 (excl. VAT).
Learning Objectives	Master 3D point cloud foundations: Understand LiDAR sensing, 3D representations, datasets and the Python toolchain. (M1, M2) Engineer detection pipelines: Apply RANSAC, KD-Tree, PCA, DBSCAN and 3D bounding-box extraction. (M3, M4) Ship detection apps: Build a 3D Python app, integrate SAM 3D and evaluate with proper metrics. (M5, M6)
Target Audience	Engineers and researchers building 3D object detection pipelines for autonomous driving, robotics and ADAS.
Prerequisites	Working Python notions help. Watch the prerequisites primer →
Estimated Duration	Approximately 22 hours of focused work. Fully asynchronous.
Access	Direct enrolment via the 3D Geodata Academy. A 14-day legal cooling-off period applies.
Accessibility & Disability	All courses are open to learners with disabilities. A dedicated referent reviews each request to put the right pedagogical and technical adjustments in place. Referent: Dr. Florent Poux — howto@learngeodata.eu.
Contact	Dr. Florent Poux — howto@learngeodata.eu 3D Geodata Academy

A note from Dr. Florent PouxThis is the engine I rebuilt three times before getting it right. The course gives you the final architecture without the dead ends — including the SAM-3D integration that took me months to wire up cleanly.

2. Technical Stack & Pedagogical Means

Software stack: Python and the standard 3D ecosystem (Open3D, Laspy, NumPy, PyTorch, CloudCompare, Meshroom — adapted to the course focus).
Hardware: any modern laptop (Windows, macOS, Linux). GPU recommended for deep learning modules.
Datasets: reference datasets and notebooks provided.
Infrastructure: proprietary LMS with 24/7 access, automated quizzes, progress tracking and digital course materials.
Modality: alternates theory and hands-on practice — code, real datasets, software use and shippable project deliverables.
Theory resources: video lessons, written handouts, curated research articles and PhD-level references.

3. Course Structure

Module	Title & Focus
M1	Foundations Course foreword, LiDAR sensing, environment setup.
M2	3D Maths & Tooling Maths, NumPy, Matplotlib, Open3D, datasets.
M3	Pre-processing & Sampling Point cloud in Python, voxel-grid sampling.
M4	Detection Algorithms RANSAC, DBSCAN, KD-Tree, PCA, bounding boxes.
M5	3D Python Apps Google Colab, n-order RANSAC, integrated workflows.
M6	Operational Workflows & SAM 3D Deep learning, photogrammetry, SAM 3D.

Why this structure, Dr. Florent PouxThe 6 modules are self-contained, so you can pick and choose. If you only have one evening, do the module that maps to your current project. The others will still be there next week.

M1 — Foundations

Course foreword, LiDAR sensing, environment setup.

Course foreword and system architecture.
LiDAR sensing fundamentals.
Anaconda environment setup.
Deep dive with Spyder IDE.
3D point cloud libraries.

M2 — 3D Maths & Tooling

Maths, NumPy, Matplotlib, Open3D, datasets.

Maths for 3D, made easy.
NumPy basics; Matplotlib for 3D; Open3D basics.
ADAS available 3D datasets.
Clustering and unsupervised techniques.
3D point clouds and representations.
Code and data access.

M3 — Pre-processing & Sampling

Point cloud in Python, voxel-grid sampling.

Point cloud in Python.
Point cloud pre-processing fundamentals.
Voxel-grid sampling.

M4 — Detection Algorithms

RANSAC, DBSCAN, KD-Tree, PCA, bounding boxes.

RANSAC in-depth and 3D ground detection.
DBSCAN and HDBSCAN.
3D data structures and KD-Tree search.
PCA in-depth and PCA for feature extraction.
3D bounding-box extraction.
K-NN clustering.

Mid-course checkpoint, Dr. Florent PouxWhen you reach M4 — Detection Algorithms, stop and apply what you've learned to a dataset you actually care about. The back half of the course goes faster when the first half sits on a real example, not a toy one.

M5 — 3D Python Apps

Google Colab, n-order RANSAC, integrated workflows.

Introduction to Google Colab.
n-order RANSAC: domain transfer.
3D integrated workflows.
3D Python app.

M6 — Operational Workflows & SAM 3D

Deep learning, photogrammetry, SAM 3D.

Coder track: 3D deep learning.
Engineer track: point cloud processing.
On-site data: 3D photogrammetry.
3D object detection and operational workflow.
3D metrics and evaluation.
Segment Anything for 3D point clouds (full pipeline).
Course outro and engine roadmap.

Expert tip — Dr. Florent PouxSkip the Colab module if you have a local GPU. Keep going if you don't — Colab is still the fastest way to learn deep learning without buying hardware.

4. Assessment, Certificate & Grading

This course is self-paced and hands-on. There is no quiz, no project to defend, no oral examination and no completion certificate. What you take away is the engine you build: the Python scripts, the results you export and the pipeline behind them.

How you check your own progress

Run each lesson's code on the dataset provided, then run it again on data of your own. If it holds on your data, you have the lesson.
Each module builds one piece of the pipeline, so a module is done when its piece runs end to end.
If something breaks, write to me at howto@learngeodata.eu with the error and the lesson number.

Accessibility & disability: the course materials can be adapted (alternative formats, extended support) on request to the disability referent howto@learngeodata.eu.

5. Course Results & Quality Indicators

3D Geodata Academy publishes its course performance indicators transparently. Figures below cover this course and are updated at the end of each session.

Indicator	Current Result	Target
Number of enrolled learners	Data being consolidated	Continuous growth
Satisfaction rate	Data being consolidated	> 95 %
Completion rate (course finished)	Data being consolidated	> 85 %
Drop-out / interruption rate	Data being consolidated	< 5 %
Recommendation rate	Data being consolidated	> 90 %

Indicators consolidated from course completions and end-of-course satisfaction surveys. Last update: April 2026.

6. Next Step

This course gives you the operational base. To go further with structured mentorship and a wider curriculum, secure your spot below or join the 3D AI Accelerator.

📅 Book a call with Dr. Florent Poux 🚀 Join the 3D AI Accelerator

The 3D AI Accelerator adds direct mentorship with Dr. Florent Poux, full access to the complete course library (20+ courses), monthly analytics on the 3D spatial AI ecosystem, curated research papers and the private job board with reviews and notes on which roles are worth pursuing.

7. Legal & Accessibility

Digital accessibility

3D Geodata Academy is committed to making its content accessible to learners with disabilities. A dedicated referent oversees content accessibility (captioning, transcripts, screen-reader friendly layouts, alternative evaluation formats). Contact: howto@learngeodata.eu.

Access for learners with disabilities

Every course can be adapted upon request. A questionnaire at enrolment captures the pedagogical and technical adjustments needed. Disability referent: Dr. Florent Poux — howto@learngeodata.eu.

Open-source licence & intellectual property

This course is made freely available for educational use. The materials remain the intellectual property of 3D Geodata Academy and Dr. Florent Poux. Reproduction or redistribution for commercial use requires prior written consent.

Data protection (GDPR)

Personal data is processed for the sole purpose of delivering the course and tracking progress. Access, rectification or deletion: howto@learngeodata.eu.