Kadr

3D environments for robots and people.█

Kadr (cadre, кадр) is a project that aims to turn real places into high fidelity 3D environments. A scan from Kadr produces two things:

One is a mesh that can be imported to NVIDIA Isaac Sim and then be used for the simulation with realistic physics.

Two is a walkable copy of the place. Imagine Google Street View, but in full 3D and ready for the Vision Pro.

The first instrument, Kadr-1, is a lidar and a global shutter camera assembled into one calibrated rig. The log below is updated as I go, and I'll do my best to publish the process and the things I learn. DM me on X or drop me an email.

Some sensors, a Jetson and a dream.
Some sensors, a Jetson and a dream.

Kadr is a weekend project. It is open-ended, and the goal is the process itself. We'll explore various sensor combinations and run into technical challenges such as time synchronization, sensor calibration, occlusion, and making sensor fusion work. Besides that, we'll touch on 3D graphics, Gaussian splats and meshes, geometry-grounded simulation, and more.

Take a look at the build log below to get started.

Build log

  • Jul 24, 2026

    We are starting! I've recently become amazed at how much leverage LLMs can give you to take an experimental project like this off the ground.

    I don't know much about CAD. I don't write firmware or drivers. Building the infrastructure for the project sounds scary. Time synchronization, frame transforms, zero-copy transport and other daily problems of a project like this are pretty daunting for one person to handle.

    Luckily, I've got three things at my disposal: (1) a general understanding of how pieces fit together, (2) the ability to manage the scope, (3) a modern tech stack with LLMs / AI as part of it. So — let's see where the project ends up.

Get updates via email. Once in 2 weeks, curated by me.