tl;dr: Gamma Camera is my open source flexible camera array project and you can build your own with https://github.com/robrussell/gamma. I made a thing I call Gamma Camera. What is it? It’s easiest to explain starting from my motivation. For the past few years I’ve been tinkering with a lot of commodity camera parts and single board computers like the Raspberry Pi and Jetson Nano. I’ve also spent a lot of time trying to get a better handle on algorithms like structure from motion (SfM) and, more recently, neural radiance fields (NeRFs).

My most recent designs used OpenSCAD. Before that I was using FreeCAD for a while. I think I’m coming back to Blender now. I love all these open source tools and I’ll get around to each one again. Blender was my go-to tool for a while to create models for 3d printing from 2017-2020. Here are the kinds of things I designed back then. Beezus Warthog (my second robocar iteration) was designed in Blender So was my next one, Henry Technical.

Last year I wrote about loading and saving video in a Jupyter Notebook for frame-by-frame processing with PyTorch. Today I’d like to explain more of the actual image processing that motivated me. It started from tinkering with Apple’s Depth Pro model. I just wanted to see how it performed with some arbitrary video and maybe use it to separate the background from the foreground. Today I’ll focus on just two main tasks that differ from the last notebook:

Over the summer I built some Slime VR trackers for myself. It’s now December and I still haven’t used them very much personally. I wrote about the part selection at the time but I never explained how to put firmware on the trackers. SlimeVR is written with the ESP32 family of microcontrollers in mind. I chose a board with the very capable ESP32-S3 for my build. There are always slight variations across different microcontrollers, even with similar names, and even from the same vendor.