This new WebGL demo demonstrates the drawing of a triangular zen doodle. Zen doodles are usually small drawings of arbitrary patterns drawn to relax or think.
The demo uses a simple triangle pattern which is created by starting with a triangle. Then, for a certain amount of times, a new triangle is created. Each new triangle (child) is rotated and then scaled to fit within the previously generated triangle (parent). There are other ways to create this pattern but this one very simple. To make things pretty, colors get more saturated to the inside and are animated. The vertices move and change size using some sine waves and the amount of rotation is also animated, giving the image a slow breathing feeling.
From the technical side, figuring out the scaling amount after each rotation is a bit tricky. To make it a bit easier I simply used triangles where all sides have the same length, which allows me to use a uniform scale. To calculate the scaling amount, the intersection of the direction vector (=line from the origin to the rotated first vertex of the parent triangle) with the right side (assuming clockwise rotation) of the parent triangle is calculated. Since everything is calculated relative to the origin, the new scale is given by the length of the intersection vector.
This small demo shows the rendering of the earth using HTML5 and WebGL.
The demo uses shaders for per pixel lighting, procedural clouds, atmosphere and a day and night cycle. The high resolution textures are taken from NASAs visible earth gallery. You can move using the WASD keys and change the view by clicking and dragging.
You can try it by clicking on the following link (WebGL-compatible browser required, e.g. Firefox, Chrome, Opera):
The rendering of transparent objects is not an easy task when using z-buffering. The reason is that the way z-buffering works requires to draw all transparent objects from back to front. However, this is not always possible since, for example, two triangles (or objects) could intersect each other, making it impossible to tell which one is in front of the other. To solve this problem, you need an algorithm which is order independent, hence the name order independent transparency. There are already a few algorithms for this: Using an A-Buffer, Depth Peeling , Screen-Door-Transparency and Stochastic Transparency, which I implemented as a test for another demo.
When I had a day off, I thought it would be fun to write a small viewer for LIDAR point cloud data.
Most of the LIDAR data which is freely available in a simple ASCII format or in the LAS file format. While it is quite easy to read the point data from both formats, the LAS format is more flexible and requires more implementation effort.
After writing the reader I tried to render aerial LIDAR point data directly, but it didn’t work as expected. The reason: The file format stores an x, y and z coordinate, but it doesn’t say in which coordinate system. Most of the time, cartographic coordinate systems are used. Fortunately, there is the PROJ4 which allows you to project the point data to a coordinate system which is more useful vor visualization. Since I’m more interested in terrestial LIDAR data, I decided to implement PROJ4 later.
There isn’t much terrestial LIDAR data available for free, however, the UC Santa Barbara provides terrestial LIDAR scans of their campus in the ASCII format. Some data sets are locally aligned, making them well suited for direct rendering.
The following screenshots show my first results, with three data sets which contain about 11.2 million points in real time:
I’m not sure if I continue this project, but the next steps would be to see how many more points can be rendered in real time, to implement a spatial hierarchy and to evaluate other rendering techniques.
So what’s new for 2013? As you can see, I migrated the old page to WordPress, giving the site a modern layout and making it easier for me to post new stuff. Since making new computer graphics demos with state of the art graphics becomes more and more time consuming, I’ll post less of them and try to focus on new articles and tutorials in the future.