Abstract
Volume rendering is an important area of study in computer graphics, due to its application in areas such as medicine, physic simulations, oil and gas industries, and others. The main used method nowadays for volume rendering is ray casting. Nevertheless, there are a variety of parallel APIs that can be used to implement it. Thus, it is important to evaluate the performance of ray casting in diferent parallel APIs to help programmers in selecting one of them. In this paper, we present a performance comparison using OpenGL® with fragment shader, OpenGL® with compute shader, OpenCL, and CUDA.
Highlights
The visualization of volume datasets has gain importance in the last decades in areas such as medicine, physic simulations, oil and gas industries, and others
The results of the execution time in milliseconds per frame is presented in Table 3 and Figure 7 for the fragment shader, compute shader, OpenCL, and CUDA
Comparing the best results between ray/box intersection test (RB) and R for each parallel API, and the fragment shader, we found that compute shader has the fastest results among all the implementations, with the exceptions of the male 128 × 128 × 311 with respect to fragment shader and CUDA, and the beetle 416 × 208 × 247 with respect to fragment shader
Summary
The visualization of volume datasets has gain importance in the last decades in areas such as medicine, physic simulations, oil and gas industries, and others. Ray casting is currently the standard in volume visualization, due to its parallel nature and rendering quality. Since this method is highly parallel, GPUbased ray casting is an attractive option to implement it with interactive frame rates and low cost. With the entry and exit point of one ray into the volume, the problem is commonly reduced to sampling the volume at constant steps, classifying the samples and composing them. It is accompanied with acceleration techniques like early ray termination [4]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have