Author: Peter Trier Mikkelsen

Here is small tech demo of our CUDA implementation of the SURF descriptor and matching method for finding a correspondence between images.. Here we are tracking an image of a number plate in a live web cam feed. Notice that the frame rate is bottlenecked by the screen grabbing used to create the video and not the object tracking.

We just updated our smoke and density field viewing software, so it runs a lot faster and we are at this moment experimenting with realtime indirect photontracing and gpu kd construction. Life is good 🙂

Here is a little movie showing real time simulation of non-linear elastic material properties using the Total Lagrangian Explicit Dynamic FEM. Three different sets of material parameters were used. Our implementation is done in CUDA. Thanks to Brian Bunch Christensen and Jens Rimestad for cooperation on the implementation. The source code is under the LGPL licence and can be found here. Please acknowledge if you use it for your research. Thanks to Movania Muhammad Mobeen for tidying up the project so it…

Welcome to my little cuda ray tracing tutorial, and first  a warning: Ray tracing is both fun and contagious, there is a fair chance that you will end up coding different variants of your ray tracer just to see those beautiful images. So if you value your spare time stop reading 🙂 I am not an ray-tracing expert, and i also value my spare time. But i really like to synthisize images on my computer, and i cannot explain why…

Here is a tech demo of our Cuda smoke visualizer software. The software demonstrates real-time interaction and visualization with a smoke data set. It is possible to adjust several parameters such as density and lighting position. To download press HERE.

Our first attempt to use photonmapping in our Cuda raytracer, 300 samples pr pixel, 1200×800, final image rendering time 65 secs. Here is the famous sponza scene in a 400 pixel pr sample rendering, without and with photonmapping. At this moment this takes a few minutes to render, but we hope to improve on the rendering time when we learn a little more about CUDA.