From Photos to Real-Time 3D
Train Gaussian Splatting models from your own photos. Optimize quality. Extract meshes. Deploy photorealistic 3D scenes that render in real time.
See Gaussian Splatting in action
From a set of photos to a photorealistic 3D scene you can fly through in real time.
Faster rendering than NeRF. Real-time framerates on consumer GPUs.
Students trained worldwide across 80 countries.
Built on the latest Gaussian Splatting research and production techniques.
Splatting tutorials get you 80%. Production needs 100%.
You have seen the demos. You ran the original 3DGS paper code. Maybe you trained a model on a few photos and got an impressive result. But when a client asks for a production-quality reconstruction of a 500 m2 building, the result looks noisy, has floaters everywhere, and the mesh extraction fails.
The gap is not running the code. It is understanding the theory deeply enough to debug quality issues, optimize training, pick the right hyperparameters, and extract usable geometry from the splat representation.
NeRFs were a breakthrough but too slow for production. Gaussian Splatting delivers the same photorealistic quality at real-time framerates. In 2024, every major 3D company is adopting it. The engineers who understand how it works, not just how to run the code, will define how 3D content is created for the next decade. This course gives you that understanding.
What you’ll build
From theory to deployed real-time 3D scenes.
Train splat models from scratch
Understand the math behind 3D Gaussians. Train models from your own photos with proper optimization. Debug quality issues like floaters, blur, and noise.
SuGaR mesh extraction
Convert splat models into clean triangle meshes using SuGaR. Get usable geometry for CAD, BIM, and game engines from neural reconstructions.
Real-time deployment
Deploy your splat scenes for real-time viewing. WebGL viewers, custom renderers, and client-ready delivery of photorealistic 3D.
Deep theory
Not just code. Understand the splatting math, the optimization landscape, and why certain scenes fail. Debug like an expert.
NeRF comparison
Hands-on comparison with NeRF. When to use splatting, when NeRF still wins, and how to make the right call for your project.
Quality metrics
PSNR, SSIM, LPIPS. Quantitative evaluation of your reconstructions. Know when your model is production-ready and when it needs more training.
I have been researching and teaching 3D reconstruction for over a decade. Gaussian Splatting is the most exciting development in the field since the invention of dense stereo matching. This course distills everything I have learned applying it to real projects into a structured curriculum you can follow in a weekend.
How this course works
Focused, hands-on, and production-ready.
100% asynchronous
Access everything 24/7 on the LMS. Self-paced. Start today, finish this weekend.
Code-along projects
Complete Python source code and training scripts. Clone the repo and start splatting immediately.
Real datasets included
Indoor scenes, outdoor environments, and architectural captures. No need to shoot your own data (but you can).
Theory + practice
Not just running the code. You will understand the math well enough to debug, optimize, and extend the pipeline.
Lifetime access
One payment, permanent access. Every update included.
Upgrade path
This module is part of the Reconstructor OS. Upgrade anytime and your payment applies as credit.
This is a focused, single-module course on Gaussian Splatting. If you want the complete reconstruction stack (photogrammetry + mesh + splatting + AI vision + deployment), consider the Reconstructor OS instead. Your purchase here applies as credit if you upgrade later.
The Curriculum
10 lessons. From theory to real-time deployment.
Prerequisites
This course requires basic Python skills and a CUDA GPU for training.
- Python (mid-level): comfortable with classes, NumPy, virtual environments
- Hardware: CUDA GPU with 8+ GB VRAM required for training. 16+ GB RAM
- Basic 3D concepts: understanding of coordinate systems and camera models helpful but not required
No prior Gaussian Splatting or NeRF experience needed.
Understand the 3D Gaussian representation, the rendering pipeline, and the training loop. Set up your environment and train your first model.
Optimize training for quality and speed. Understand adaptive density control, learning rate scheduling, and how to debug common artifacts.
Extract clean triangle meshes from trained splat models using SuGaR. Get usable geometry for downstream applications.
Hands-on comparison with NeRF. When splatting wins, when NeRF still has advantages, and how to choose.
Deploy your splat scenes for real-time viewing. Custom renderers, viewer integration, and client-ready delivery.
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 and researchers working with 3D reconstruction.
“The Gaussian Splatting module alone was worth the price. Florent breaks down the math in a way that makes sense for engineers, not researchers.”
“Debugging splat artifacts used to be guesswork. After the training optimisation module, I can reason about density, learning rates, and floaters like an adult. Huge leap.”
“The SuGaR mesh extraction workflow is what finally made splatting viable for our CAD deliverables. Clean geometry, clean pipeline, clean handover.”
“The deployment module changed how I think about delivering 3D data. We went from emailing .las files to streaming point clouds in the browser. Clients couldn’t believe it.”
Get lifetime access
One payment. Every lesson, every update, every line of code.
Gaussian Splatting for 3D
Complete Gaussian Splatting module + source code + datasets + lifetime updates
- 10 hands-on lessons (12+ hours)i
- Complete Python source code + datasets
- SuGaR mesh extraction pipeline
- NeRF comparison framework
- Lifetime access + all future updatesi
- 90-day results guaranteei
Zero-risk guarantee: If you don’t see real results within 90 days, full refund. 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 | Gaussian Splatting for 3D | Course Library | 3D AI Architecti | Enterprise |
|---|---|---|---|---|---|
| Courses included | 1 topic | 10 lessons | Full catalogi | 3 OS courses + Library (tiered) | Custom |
| Hours of content | 2-8h | 12+ 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 | €297 | €1,297 | Starts at €1,999 | On request |
Straight answers
Do I need prior Gaussian Splatting experience?
No. I start from the theory and build up to training and deployment. You need Python and a CUDA GPU.
What GPU do I need?
NVIDIA GPU with 8+ GB VRAM. RTX 3060 or better recommended. Training is GPU-intensive. Rendering works on lower-end hardware.
How is this different from the free 3DGS repo?
The paper code gets you a trained model. This course teaches you why it works, how to optimize it, how to debug quality issues, and how to extract usable geometry and deploy it. Production skills, not just a demo.
Can I upgrade to the Reconstructor OS later?
Yes. Your purchase applies as credit toward the Reconstructor OS. Contact me when you are ready to upgrade.
How long do I have access?
Lifetime. One payment, permanent access. Every future update included.
What’s the refund policy?
90 days. Apply the material and train a model. If you are not satisfied, email me for a full refund.
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