英语论文网

Stanford Tech Report CTSR 2005-02
Light Field Photography with a Hand-held Plenoptic Camera
美国斯坦福大学技术报告Ren Ng∗ Marc Levoy∗ Mathieu Br´edif∗ Gene Duval† Mark Horowitz∗ Pat Hanrahan∗∗Stanford University †Duval Design

Abstract
This paper presents a camera that samples the 4D light field on itssensor in a single photographic exposure. This is achieved by insertinga microlens array between the sensor and main lens, creatinga plenoptic camera. Each microlens measures not just the totalamount of light deposited at that location, but how much light arrivesalong each ray. By re-sorting the measured rays of light to
where they would have terminated in slightly different, syntheticcameras, we can compute sharp photographs focused at differentdepths. We show that a linear increase in the resolution of imagesunder each microlens results in a linear increase in the sharpnessof the refocused photographs. This property allows us to extendthe depth of field of the camera without reducing the aperture, enablingshorter exposures and lower image noise. Especially in themacrophotography regime, we demonstrate that we can also computesynthetic photographs from a range of different viewpoints.These capabilities argue for a different strategy in designing photographicimaging systems.To the photographer, the plenoptic camera operates exactly likean ordinary hand-held camera. We have used our prototype to takehundreds of light field photographs, and we present examples ofportraits, high-speed action and macro close-ups.

Keywords: Digital photography, light field, microlens array, synthetic photography, refocusing.

1 Introduction
Conventional cameras do not record most of the information aboutthe light distribution entering from the world. The goal of the camerapresented in this paper is to re-capture this lost information:to measure not just a 2D photograph of the total amount of light ateach point on the photosensor, but rather the full 4D light field measuringthe amount of light traveling along each ray that intersectsthe sensor. One can also think of this as capturing the directionallighting distribution arriving at each location on the sensor.
The purpose of capturing the additional two dimensions of datais to allow us to apply ray-tracing techniques to compute syntheticphotographs flexibly from the acquired light. The overall conceptis to re-sort the rays of light to where they would have termninatedif the camera had been configured as desired. For example, wedemonstrate that we can shoot exposures with a relatively large f/4lens aperture (for a short exposure time and low image noise), andyet compute photographs where objects at any depth are as sharpas if taken with a relatively small f/22 aperture. This result is away of decoupling the traditional trade-off in photography betweenaperture size and depth of field.
Externally, our hand-held light field camera looks and operatesexactly like a conventional camera: the viewfinder, focusing mechanism,length of exposure, etc. are identical. Internally, we augmentthe 2D photosensor by placing a microlens array in front ofit, as proposed by Adelson andWang [1992] in their work on the“plenoptic camera” (They did not build this device, but prototypeda non-portable version containing a relay lens.) Each microlensforms a tiny sharp image of the lens aperture, measuring the directionaldistribution 本论文由英语论文网提供整理，提供论文代写，英语论文代写，代写论文，代写英语论文，代写留学生论文，代写英文论文，留学生论文代写相关核心关键词搜索。