Abstract
In this paper, we present a new approach for collision detection of volumetric objects. Good speeds and accuracy can be achieved using existing graphics hardware. The algorithm is simple to implement and, unlike other collision detection mechanisms, requires very little extra memory and preprocessing time. Using a combination of the two-dimensional scanline process and three-dimensional texture mapping technique, a number of polygonal slices can be computed for all volume objects in a scene. For each slice, the objects are checked for collisions using the framebuffer and logical color coding operations. Experiments are provided and analyzed.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have