Study of depth of field based on GPU post-processing

Computer generated images by default render the entire scene in perfect focus. And computer graphics cameras lack the finite depth of field present in real world, this result in reducing the realism of the scene. Several depth of field methods have been proposed, in this paper, object space methods depth of field will be introduced, and the ray-racing method will be put as the representative method. After then, this paper will focus on post-processing method based on GPU. The comparison result will be shown as the experiment section.