Rafael Brune

Currently, a Senior Sensor Simulation Specialist at NVIDIA.

Previously, a Senior Architect for raytracing based physically accurate radar, lidar and camera simulation at AAI. For AAI I developed from the ground up and maintained a raytracing based physically accurate sensor simulation framework.

Before that, a Research and Development Engineer at colormass. There I developed, built and maintained the hardware and software of a SV-BRDF capture system for the digitization of large material samples. The system includes automatic repeating pattern detection, seamless tiling and fitting of the raw data to PBR models for accurate material reproduction in modern rendering engines.

mail | github | twitter | linkedin

Computational Neuroscience

Theoretical Neuroscience Group, Max-Planck-Institute for Experimental Medicine

Margin Learning in Spiking Neural Networks

R. Brune and R. G├╝tig
[Paper and PhD thesis to be published]

Dynamics on Complex Networks

Research on Complex Systems Group, Northwestern University
Nonlinear Dynamics Group, Max-Planck-Institute for Dynamics and Self-Organization

Modularity maximization and tree clustering: Novel ways to determine effective geographic borders

D. Grady, R. Brune, C. Thiemann, F. Theis, D. Brockmann, Handbook of optimization in complex networks, Springer US 2012

The Structure of Borders in a Small World

C. Thiemann, F. Theis, D. Grady, R. Brune, D. Brockmann, PLoS ONE 2010

Computational quantitative projections for H1N1 flu dynamics in the United States 2009

Designed and implemented a fully stochastic spatial-temporal reaction-diffusion algorithm used to simulate a disease spreading model. Later on optimized the algorithm with inline assembly for the PowerPC architecture of the IBM Blue Gene supercomputer run by the Argonne National Laboratory.



BTCsim is a stochastic event-based continuous-time Bitcoin network simulator. It can be used to study block chain generation, forking, block propagation, network dynamics, miner/mining-pool interactions/strategies and effects of blocksize on the network.

Art and data visualization

3D printed cityscapes from LIDAR data

Using freely available high resolution LIDAR data and GIS tools to generate a height map of the target region. This height map is then converted into an STL model that can be used with 3D printers.
[coming soon]

Valve Lighthouse / HTC Vive tracking volume visualizer

Visualization tool for Valves Lighthouse infrared laser based tracking technology. It shows the size and goodness of the tracking volume based on direction and position of the two base stations.

Hardware hacking and reverse engineering

Magic Lantern

3×3 raw video recording for the Canon EOS-M DSLR camera.
https://www.magiclantern.fm [Use crop_rec module for 3×3 raw]

Calculator for Magic Lantern raw video resolution, bandwidth and crop.


Voltage regulator module monitoring and control via I2C and PMBus for AMD GPUs.
VRMtool v1 [Closed source, C++, release thread on overclock.net]
https://github.com/rbrune/VRMtool [Open Source, Python, in progress v2]

Nike FuelBand fitness tracker

To allow for interoperability with Android devices and Linux I reverse engineered the USB-HID protocol used by the device, wrote a python script to interact with it as well as a proof of concept Bluetooth Android app to sync step data to Google Fit. Decryption and analysis of full firmware not quite done yet.
https://github.com/rbrune/fuelband-usb [Python USB-HID script]
https://github.com/rbrune/Fuelbandsync [Android App to sync via Bluetooth and upload to Google Fit]

Motorola Lapdock EDID patching

[coming soon]

Android custom ROM development

Google Nexus Q as an USB DAC+AMP

This ROM turns your Nexus Q into an USB audio card. When you turn on the Nexus Q it will register as an USB audio device in the USB host Linux/OSX/Windows/Android system. You will be able to listen to music and other sounds through the speakers connected to the Nexus Q internal amp.

HP Touchpad

As part of the CyanogenMod team I helped getting Android up and running on the HP Touchpad. My contributions include a from scratch implementation of the touchscreen driver as well as minor changes for device upbringing.

B&N Nook Color

After acquiring a Barnes & Nobles Nook Color E-Reader I got involved with the CyanogenMod team working on porting Android 2.3 to the device. I then used the Android 3.0 Honeycomb preview emulator images, some of the B&N modules+driver, pieces from the upcoming CyanogenMod 7 and manually patched ARM assembler in libraries, wrote some scripts to hack around various issues and spent a lot of time – to have the first device running Android 3.0 outside of Google.
https://www.youtube.com/watch?v=cKQHZTihjrU [Video of Android 3.0 on Nook Color]
https://forum.xda-developers.com/showthread.php?t=936995 [Release of Android 3.0 for Nook Color]
https://www.engadget.com/2011/01/28/nook-color-earns-its-very-eary-very-unofficial-android-3-0-hone/ [Engadget coverage]
http://www.zdnet.com/article/android-3-0-honeycomb-download-comes-to-nook-color/ [ZDnet coverage]


Python programming and data analysis course for PhD students