Build the Full Stack of 3D Deep Learning
From your first ANN to a deployed 3D AI app. 7 production modules. Every architecture that matters: PointNet, PointNet++, KPConv, GrowSP. Plus generative, hybrid, and LLMs for 3D. No black boxes. Just code you own.
See the complete stack in action
A walkthrough of every module, every architecture, and the client-server AI app that ties them together.
Points per scene. Segmented by production PointNet++ and KPConv pipelines you build.
Students trained worldwide across 80 countries.
Deep learning research and production experience distilled into structured workflows.
You know a bit of everything. You ship nothing.
You’ve trained a PointNet. You’ve read the KPConv paper. You’ve seen a demo of a generative 3D model on Twitter. You can run a tutorial. You can even tweak a config. But when your client asks for a production 3D AI system, you stall. There are too many pieces. They don’t fit together yet.
That’s the OS gap. An operating system isn’t a pile of scripts. It’s a modular, interconnected framework where every component talks to the others. Most 3D deep learning courses teach components in isolation. This one teaches the complete stack: from ANN fundamentals through generative 3D to a full client-server AI app. Because that’s what production requires.
An Operating System is the infrastructure other systems run on top of. This program is built the same way. Each module installs a layer: fundamentals, image networks, point cloud networks, advanced architectures, production scaffolding, generative and hybrid intelligence, and the client-server deployment path. You don’t just finish with skills. You finish with a reusable framework you can deploy across every 3D AI project you touch for the next decade. The stack, not the snippets.
What you’ll build
Not exercises. A full 3D deep learning stack you deploy from scratch.
Production segmentation
The complete semantic segmentation pipeline on real aerial LiDAR. PointNet, PointNet++, KPConv, GrowSP, side by side.
Client-server AI app
A deployed client-server application that runs 3D inference on demand. API, data contract, and delivery all hand-built.
Generative + hybrid 3D
Generative models for 3D. Hybrid systems mixing algorithmic and learned components. The frontier of 3D AI.
Every architecture
ANN, CNN, ResNet, EfficientNet, 3D CNN, 3D R-CNN, PointNet, PointNet++, KPConv, GrowSP, RNNs, Transformers. You code, configure, train, and deploy each one.
LLMs for 3D
How LLMs integrate into 3D pipelines. Code generation, annotation acceleration, and reasoning over 3D scenes.
Production framework
A reusable personal framework. Project structure, training step, inference step, YAML config, debug setup. Ship faster on every future project.
I’ve been building and teaching 3D deep learning for more than a decade. This OS is the distilled result. It’s the training I wish I had when I was stitching 3D networks together from isolated tutorials. It’s also the training my Fortune 500 clients paid me to deliver, restructured into a self-paced program anyone can follow.
How this OS works
Built for engineers who want the full stack, delivered by someone who’s shipped it.
100% asynchronous
Access every module 24/7 on the LMS. No live sessions required. Work at your own pace.
Code-along projects
Every module ships with complete PyTorch source code. Clone the repo, run it, modify it. No copy-pasting from videos.
Progress tracking
A built-in progress dashboard. Track completion across all modules, mark milestones, and monitor your learning velocity.
Production-first pedagogy
No toy datasets. Every exercise uses real-world data at real scale. You finish with systems that run outside a notebook.
Lifetime access
You keep access forever. Every future module, every code update, every new architecture. If anything ever happens to the platform, I’ll ship you the full offline version.
Ecosystem of standalone courses
If you already own one of the standalone courses (Foundations, Advanced Architectures, Engineering Neural Networks), your purchase applies as credit. Just email me.
Copilot writes plausible PyTorch. ChatGPT can sketch a training loop. But neither of them knows the memory budget of your GPU, the class imbalance of your dataset, or why KPConv crashes on your aerial scene while PointNet++ trains fine. Without that context, they guess. This OS teaches you to be the one directing AI: the architectural judgment, the system-level thinking, the production decisions that turn suggestions into working products. That’s the engineer who gets paid in the AI era.
The Operating System
7 modules, 50 lessons. Each module is a layer. Together, they form the complete 3D AI stack.
Prerequisites
This OS picks up where tutorials leave off. If you feel below on any point, a prerequisite ebook is included.
- Python (mid-level): classes, file I/O, NumPy, virtual environments
- Basic linear algebra: vectors, dot products, matrices. I explain the rest as we go
- Hardware: 16 GB RAM minimum, 32 GB+ recommended. CUDA GPU with 8+ GB VRAM required for KPConv, PointNet++, and GrowSP
- Software: Python, PyTorch, Open3D, NumPy, Matplotlib. All free and open source
No prior 3D deep learning experience required. I start at the ANN and build all the way up.
A structured orientation to 3D deep learning. Data types, workflows, and the first Python setup. The shared base every other module rests on.
Build every core architecture from scratch. ANN, CNN, ResNet, EfficientNet. Plus RNNs and Transformers on demand.
The 3D-specific data layer. Voxel datasets, PyTorch Dataset classes, and the preprocessing that makes or breaks training.
The PointNet family end to end. PointNet and PointNet++: fundamentals, data prep, model creation, training, inference, evaluation.
The production-grade architectures. 3D CNN, 3D R-CNN, KPConv, and GrowSP. System design thinking and code structure.
The frontier of 3D AI. Generative models, hybrid systems, and how LLMs integrate into 3D pipelines.
Ship. Workflow and system design for production. Step-by-step 3D Python app production. Client-server AI app: setup, build, debug, finalization. The deployed artifact you can show clients.
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, researchers, and professionals shipping 3D AI in 80 countries.
“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.”
“Implementing PointNet from scratch alongside Florent finally made the math feel like code. I caught a tensor-shape bug in our internal model the next day.”
“The PointNet++ to KPConv progression is taught the way I wish someone had taught me five years ago. Each architecture justifies the next, no hand-waving.”
“I needed to choose between sparse convolution backbones for a transit project. The architecture comparison module gave me a defensible recommendation in two days.”
Get lifetime access
One payment. Every module, every update, every line of production code.
3D Deep Learning OS
Full production curriculum + source code + lifetime updates
- 7 production modules (50+ hours)i
- Complete PyTorch source code
- PointNet, PointNet++, KPConv, GrowSP
- 3D CNN, 3D R-CNN, generative, hybrid, LLMs
- Client-server AI app project
- Lifetime access + all future updatesi
- 90-day results guaranteei
Zero-risk guarantee: Apply the course material. If you don’t see real results within 90 days, I’ll refund you in full. No forms, 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 | 3D Deep Learning OS | Course Library | 3D AI Architecti | Enterprise |
|---|---|---|---|---|---|
| Courses included | 1 topic | 7 modules | Full catalogi | 3 OS courses + Library (tiered) | Custom |
| Hours of content | 2-8h | 50+ 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 | €1,497 | €1,297 | Starts at €1,999 | On request |
Straight answers
Do I need prior deep learning experience?
No. You need mid-level Python and basic linear algebra. I start at the ANN and build every architecture up from there. If you want a gentler on-ramp to neural networks, start with Engineering Neural Networks for Geospatial Analysts.
What hardware do I need?
Minimum 16 GB RAM. For KPConv, PointNet++, and GrowSP modules, 32 GB+ and a CUDA GPU with 8+ GB VRAM are required. Most other modules run on a modern laptop.
How long do I have access?
Lifetime. One payment, permanent access. Every future update included. No subscriptions, no expiration, no hidden fees.
What’s the refund policy?
90 days. Train real models on real data. If you don’t see results, email me and I’ll refund you in full. No forms, no questions.
How is this different from the standalone courses?
The OS is the complete production path. The standalones (Foundations, Advanced Architectures, Engineering Neural Networks) are focused modules extracted from it. The OS adds generative 3D, hybrid systems, LLMs for 3D, and the full client-server AI app that the standalones don’t include.
I already own a standalone course. Can I upgrade?
Yes. Your purchase applies as credit toward the OS. Email me at howto@learngeodata.eu with your receipt and I’ll arrange it.
Can AI replace what this OS teaches?
AI writes plausible PyTorch. It suggests architectures. But it doesn’t know your GPU’s memory budget, your dataset’s class imbalance, or why KPConv is the right call for your aerial scan. Without that context, AI guesses. This OS teaches you to be the one directing AI. The architectural judgment and system-level thinking that turn AI suggestions into shipped products.
Do you offer team or enterprise pricing?
Yes. For teams of 3+ or enterprise licensing, contact me at howto@learngeodata.eu. I offer volume discounts and tailored onboarding.
Why does the price increase?
Every time I add a new module, architecture, or project to the OS, the value goes up, and so does the price. Lock your price now, get every future addition at today’s rate. The longer you wait, the more you pay for the same access.
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