Automate Scene Labeling with SegmentAnything 3D
Project the SAM foundation model into 3D space and label entire point cloud scenes automatically. 5 focused lessons, zero manual annotation.
See automated labeling in action
A raw indoor scan goes in. A fully segmented, instance-labeled scene comes out. No clicks in between.
Faster than manual annotation on typical indoor scenes. The bottleneck of every supervised 3D ML project, removed.
Students trained worldwide across 80 countries.
Production experience distilled into structured, repeatable workflows.
Your 3D ML project is stuck at the labeling stage.
Every supervised 3D pipeline starts the same way, someone has to label the data. So you sit there, lasso-selecting points, scene after scene, hour after hour. The model you actually wanted to build is still weeks away, and your enthusiasm is dying one annotation at a time.
Meanwhile, foundation models changed the rules. SegmentAnything segments any object in any image, zero-shot. The trick this course teaches: render your point cloud into images, let SAM do the segmentation, then project the masks back onto the 3D points. The scene labels itself.
Labeling is the single most expensive line item in any supervised 3D project, in money or in your own evenings. Automate it, and projects that were not viable suddenly are. This is the highest leverage-per-lesson course in the catalog.
What you’ll build
A complete automated labeling pipeline, from raw scan to segmented scene.
SAM inference pipeline
Run SegmentAnything locally on your own machine. Model setup, prompting modes, and mask generation on real images.
3D-to-2D projection engine
The geometric core. Render point clouds to images with known camera poses, so every pixel maps back to its source points.
Automated 3D labeler
The full loop. SAM masks lifted from 2D back onto the 3D point cloud, fused across views into clean per-point instance labels.
Reproducible environment
A clean, documented Python setup for SAM and the 3D stack. The same environment recipe I use for client work.
Real indoor scenes
The pipeline runs on real captured point clouds, with all their occlusions and noise. What works here works on your data.
Training data factory
The output is not just pretty colors. It is labeled training data you can feed straight into PointNet or any supervised 3D model.
Manual labeling is the bottleneck in every 3D project I have ever consulted on. This course is the workflow I built to break through it, projecting SegmentAnything into 3D space so the scene labels itself. Five lessons, one pipeline, and your evenings back.
How this course works
Short, sharp, and built around one complete pipeline.
100% asynchronous
Access everything 24/7 on the LMS. Self-paced. No live sessions required.
One-sitting format
5 focused lessons. You can have the full pipeline running on your own data within a weekend.
Code-along build
Every lesson builds a piece of the pipeline. By the end, the parts click together into one automated system.
Foundation model workflow
Learn the projection pattern once, and you can lift any 2D foundation model into 3D, not just SAM.
Lifetime access
One payment, permanent access. Every future update included.
Feeds your ML stack
The labels you generate here are the training data for every supervised course in the catalog.
Foundation models will keep improving, and every new one will be 2D-first. The 3D-to-2D-and-back projection workflow you build in this course is the permanent bridge. Swap the model, keep the pipeline.
The Curriculum
5 lessons. From SAM setup to a self-labeling 3D scene.
Prerequisites
This course is for engineers and researchers who need labeled 3D data and refuse to produce it by hand.
- Python (beginner+): comfortable with loops, functions, and basic NumPy operations
- Basic 3D knowledge: you know what a point cloud is and have opened one before
- Hardware: 16 GB RAM recommended. A GPU speeds up SAM inference but is not required
- Software: Python, PyTorch, SegmentAnything, Open3D. All free and open-source
No prior deep learning experience required. SAM is used as a tool, not built from scratch.
Get the foundation model running. Course overview, a clean Python environment, and SAM inference on 2D images with all its prompting modes.
The core of the course. Project point clouds to images, run SAM, and lift the masks back into 3D for fully automated scene labeling.
Your instructor
Dr. Florent Poux
I’ve spent 12+ years in 3D geospatial: from field surveys with total stations to building AI systems for Fortune 500 companies. I published the O’Reilly book on 3D Data Science with Python. I’ve advised startups valued at over 15M EUR. I’ve held a professorship, taught at university, and led R&D for some of the largest organizations in the space.
I don’t teach syntax. I teach judgment. Every module is built around real decisions I’ve faced in production. Which neural renderer fits an industrial inspection job. How to architect a semantic pipeline that doesn’t choke on 500M points. When to use algorithmic methods and when to switch to deep learning.
What students say
Engineers, GIS professionals, and researchers from 80 countries.
“Our team processed 200M points for a highway survey. Before this course, we were stuck at 10M with crashes. The memory management module saved us weeks of work.”
“I’m a surveyor with 20 years of field experience. This gave me the Python and AI skills to modernize our entire workflow. Best investment I’ve made in my career.”
“RANSAC plus DBSCAN on a 120M-point mining dataset — segmented, volumetric change detected, and report ready in one afternoon. That used to take a week.”
“This course is the foundation I wish I had when I started with LiDAR. The feature extraction module alone reshaped how I approach every project.”
“I came in fluent in Python but lost on 3D. After Module 2 I was reading PointNet++ papers like a normal human. The intuition Florent builds is the missing link.”
“We replaced a brittle classical pipeline with a sparse-conv network thanks to this course. Production accuracy jumped from 78 to 94 percent on the same data.”
“The training loop debugging chapter was worth the price by itself. I now know why my models stop learning instead of guessing at hyperparameters.”
Get lifetime access
One payment. The full automated labeling pipeline, forever.
SegmentAnything 3D Labeling
Complete automation curriculum + Python source code + sample scenes + lifetime updates
- 5 lessons (5+ hours, 2 modules)i
- Complete Python source code + sample data
- SAM setup + 3D projection engine
- Automated per-point labeling workflow
- Lifetime access + all future updatesi
- 90-day results guaranteei
Zero-risk guarantee: If you don’t see real results within 90 days, I’ll refund you in full. No questions.
The complete ecosystem
3D AI Architect Program
The complete spatial AI curriculum, delivered in 3 tiers. Pick the depth that matches where you are — Foundations to get moving, Professional for the full OS stack, Ultimate for live access and priority support.
- 3D AI Acceleratori: 17 episodes in 6 acts
- 3D Course Libraryi: 24+ standalone courses
- All 4 OS courses (Professional & Ultimate tiers)
- Neurones 3D software access
- Monthly drop-in sessions with Dr. Poux (Ultimate)
- Spatial AI job and market intel
- Priority support + services access (Ultimate)
- 300+ hours of content
What you’re getting access to
Everything I’ve built over 12+ years, from land surveying in the field to advising 15M EUR startups, compressed into one curriculum you can start today. Delivered by the first QUALIOPI-certified 3D geospatial academy.
Every pipeline was battle-tested on Fortune 500 projects processing billions of points. You’re getting the real playbook, not theory.
Methods validated by peer-reviewed publications, the ISPRS scientific community, and 1,500+ academic citations. Not guesswork.
Built by someone who surveyed in the field, defended a PhD, advised funded startups, and shipped products to Fortune 500 clients.
I share more free content than most people put behind a paywall. That’s intentional. I want you to know exactly what you’re getting before you invest. This course is the concentrated, structured version of everything I know. No fluff. No filler. Just the production path.
Find the right path for you
From single courses to the complete ecosystem.
| Feature | Standalone Course | SegmentAnything 3D Labeling | Course Library | 3D AI Architecti | Enterprise |
|---|---|---|---|---|---|
| Courses included | 1 topic | 2 modules | Full catalogi | 3 OS courses + Library (tiered) | Custom |
| Hours of content | 2-8h | 5+ hours | 150+ hours | 300+ hours (tiered) | Custom |
| Production source code | ✓ | ✓ | ✓ | ✓ | ✓ |
| Lifetime access | ✓ | ✓ | – | ✓ | ✓ |
| 3D AI Accelerator Tracki | – | – | – | ✓ | ✓ |
| Neurones 3D softwarei | – | – | – | ✓ | ✓ |
| Spatial AI job & market inteli | – | – | – | ✓ | ✓ |
| Monthly drop-in sessionsi | – | – | – | ✓ | ✓ |
| Priority support + services accessi | – | – | – | ✓ tiered | ✓ |
| Custom onboardingi | – | – | – | – | ✓ |
| Team licensing | – | – | – | – | ✓ |
| Price | €97 – €497 | €197 | €1,297 | Starts at €1,999 | On request |
Straight answers
Do I need deep learning experience?
No. SAM is used as a ready-made tool. You load it, prompt it, and build the 3D projection around it. No model training involved.
What software do I need?
Python, PyTorch, Meta’s SegmentAnything, and Open3D. All free and open-source. No paid licenses required.
Do I need a GPU?
SAM inference is faster with one, but the course datasets are sized to run on CPU. 16 GB RAM recommended.
Will this work on my own point clouds?
Yes, that is the point. The pipeline is built to be re-pointed at your own scenes. Indoor scans work best, outdoor scenes are discussed with their caveats.
How long do I have access?
Lifetime. One payment, permanent access. Every future update is included.
What’s the refund policy?
90 days. Run the pipeline on your own scene. If you don’t see results, email me for a full refund.
How is this different from the segmentation courses?
The segmentation courses teach models that learn from labeled data. This course produces that labeled data automatically. They are two sides of the same workflow, and this one comes first.
Can I upgrade to an Operating System later?
Yes. When you are ready for the full production stack, contact me for credit toward the 3D Deep Learning OS or the Spatial OS.
Not sure if this course fits?
If you have specific questions about how the curriculum applies to your role, your team’s needs, or your technical background, I’m happy to help you figure it out before you commit.
Book a 15-min call