Learning about Shadows from Artists

Abstract

Renaissance artists discovered methods for imaging realistic depth on a two dimensional surface by re-inventing linear perspective. In solving the problem of depth depiction, they observed how shadows project and volumes flatten in nature. They investigated how controlled illumination projects volumes onto walls, exploring the phenomena long before physical optics, such as the camera, existed. This paper specifically examines artists' constructions for depicting shadows, a 3D double projection problem that artists solved completely within two dimensions. The larger goal is to develop new computational methods for creating 3D perceptions without having to leave the 2D canvas. Those methods have potential application in constructing user interfaces, in 2D image compositing and in simultaneous 2D/3D composition. This paper develops geometric constructions for casting shadows onto planar surfaces, adapted from artists' methods. Their algebraization for integration into imaging software is demonstrated, and their optically accuracy is shown. Finally, resulting images are included, along with a discussion of limitations.