A Qualitative Expert Knowledge Approach to Rendering Optimization

The rendering process allows the developer to obtain a raster 2D image from the definition of a 3D scene. This process is computationally intensive if the source scene has a certain complexity or high-quality images are required. Therefore, a lot of time is spent and many computational resources are needed. In this paper, a novel approach called QUEKARO (standing for a QUalitative Expert Knowledge Approach to Rendering Optimization) is presented for adjusting some relevant parameters involved in the rendering process by using expert systems. This way, the developer can obtain optimized results which reduce the time spent in the rendering process and, in most cases, do not affect the final quality of the raster 2D image. These results will be exposed on the result section, in which different optimizations will be studied. As we discuss on the final section of this paper, the use of expert systems in the rendering process involves a novel approach which reduces drastically the resources used and provides us with a high-scalable system. Using these arguments, we will justify the inclusion of expert systems in this area and will study future works.