Increasing GPU-Speedup of Volume Rendering for Images with High Complexity
Sang-Soo Jun, Ok-Kyoon Ha · 2015
This paper presents a speedup improvement method for rendering complex volume images in GPU platforms. Improving GPU-speedup for complex volume images are important because previous work has shown that the speedup of volume rendering optimized with transparent voxel skipping method decreases with dependency on the complexity of target images. To improve GPU-speedup of volume rendering, we present a new algorithm, called RBDV that reduces the branch divergence in transparent voxel skipping by factoring out structurally similar code from branch paths in GPU programs. We empirically proved that this RBDV algorithm increases the GPU-speedup of transparent voxel skipping at least by 14%, improving it from x17.5 up to x20.0 or more, on average, for complex target images.