Photographic Depth of Field Blur Rendering

Abstract

The purpose of our study is to introduce a new method for depth of field blur simulation used for compositing real and synthetic images in a realistic way. Existing methods do not simulate accurately the depth of field blur produced by photographic equipment. The problem is twofold: photometric and geometric. Existing methods based on convolution do not produce a realistic photometric rendering compared to real photograph and the photographic blur contains patterns associated with the diaphragm shape, neither handled by existing methods. Our method, based on image processing, addresses these two issues by following the photographic image creation process. It can be used on synthetic or real images, essentially for special effects compositing.