Light-Field Acquisition System That Facilitates Camera and Depth-of-Field Compositing in Post-Production

Abstract

Traditional film and broadcast cameras capture the scene from a single viewpoint, while important creative parameters such as the camera position and orientation, the depth of field, and the three-dimensional (3D) parallax when using a 3D camera rig are burned into the footage during acquisition. To realize artistic effects such as the matrix or the vertigo effect with traditional cameras is a complex task that requires skilled personnel. Within the matrix effect, the scene appears frozen while a camera movement is simulated by placing dozens of cameras in a mainly horizontal arrangement. The vertigo effect requires physical movement of the camera, which is usually mounted on a dolly, and translates forward and backward to and from the scene while the zoom and possibly the focus are changed accordingly. Usually, the scene needs to be shot multiple times until the best combination of zoom corrections and dolly movement is found. Against this background, a system is presented that allows capture of the light field of a sequence and thus offers full creative leeway of changing camera paths and orientation, as well as setting the focal plane in post-production. In contrast to previous approaches, the handling of the involved acquisition system is compact and thus suitable for on-set operation. This paper describes the associated content production and presents first results that underlie the suitability of the approach.