DOC PREVIEW
Introducing Vision-Realistic Rendering

This preview shows page 1-2 out of 7 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 7 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 7 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 7 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Thirteenth Eurographics Workshop on Rendering (2002)P. Debevec and S. Gibson (Editors)Introducing Vision-Realistic RenderingBrian A. Barsky Adam W. BargteilDaniel D. GarciaStanley A. Klein Computer Science DivisionSchool of OptometryBioengineering Graduate GroupUniversity of California, Berkeley, California, 94720-1776, USAhttp://www.cs.berkeley.edu/optical/vrrAbstractWe introduce the concept of vision-realistic rendering–the generation of images that incorporate characteristicsof a particular individual’s optical system. We then describe a pipeline for creating vision-realistic images. First,a subject’s optical system is measured by a Shack-Hartmann wavefront aberrometry device. This device outputsa measured wavefront which is sampled to calculate an object space point spread function (OSPSF). The OSPSFis then used to blur input images. This blurring is accomplished by creating a set of depth images, convolvingthem with the OSPSF, and finally compositing to form a vision-realistic image. We discuss applications of vision-realistic rendering in computer graphics as well as in optometry and ophthalmology and note that our method isa post-process and can handle simple camera models as a special case.Categories and Subject Descriptors (according to ACM CCS): I.3.3 [Computer Graphics]: Display algorithms, View-ing algorithms1. IntroductionWe introduce a new concept which we call vision-realisticrendering–the generation of images that incorporate charac-teristics of a particular individual’s optical system. We de-scribe a pipeline to achieve vision-realistic rendering andshow some example images. These are the first images incomputer graphics that are generated on the basis of the spe-cific optical characteristics of actual individuals.There are two distinct impacts of this research, one fromthe point of view of optometry and ophthalmology, and theother from the perspective of computer graphics. Our tech-nique enables the generation of vision-realistic images andanimations that demonstrate specific defects in how a per-son sees. Such images could be shown to an individual’soptometrist or ophthalmologist to convey the specific visualanomalies of the patient. Doctors and patients could be edu-cated about particular vision disorders by viewing imagesthat are generated using the optics of various ophthalmicconditions such as keratoconus (Figure 1) and monoculardiplopia.One of the most compelling applications is in the con-text of vision correction using laser corneal refractive eyeFigure 1: Vision-realistic image simulating vision based onactual wavefront data from a patient with keratoconus.surgeries such as LASIK (laser in-situ keratomileusis). Cur-rently, in the United States alone, a million people per yearchoose to undergo this elective surgery. By measuring sub-jects pre-op and post-op, our technique could be used toconvey to doctors what the vision of a patient is like beforeand after surgery (Figure 5). In addition, by using modeledor simulated wavefront measurements, this approach couldcThe Eurographics Association 2002.Barsky, Bargteil, Garcia, and Klein / Introducing Vision-Realistic Renderingprovide accurate and revealing medical visualizations of pre-dicted visual acuity and of simulated vision. Potential can-didates for such surgery could view these images to enablethem to make more educated decisions regarding the proce-dure. Still another application would be to show such candi-dates some of the possible visual anomalies that could arisefrom the surgery, such as glare at night.There are also interesting applications of our technique inthe context of computer graphics and computer animation.For example, vision-realistic rendering could enhance the re-alism of a first-person view. As a special case, this approachcan be used as a post-process to simulate camera effects suchas depth of field. Note that the depth map can be manipulatedto achieve non-photorealistic focusing effects, such as keep-ing a range of depths all in perfect focus. This aspect of oursystem provides powerful control that is not available whenthe camera model is incorporated in the renderer.1.1. Previous and Related WorkThe first synthetic images with depth of field were computedby Potmesil and Chakravarty28who convolved images withdepth-based blur filters. However, they ignored issues relat-ing to occlusion, which Shinya31subsequently addressed us-ing a ray distribution buffer. Rokita29achieved depth of fieldat rates suitable for virtual reality applications by repeatedconvolution with 33 filters and also provided a survey ofdepth of field techniques30. Stochastic sampling techniqueswere used to generate images with depth of field as well asmotion blur by Cook et al.6, Dippe and Wold7, and Lee etal.16. More recently, Kolb et al.15described a more com-plete camera lens model that addresses both the geometryand radiometry of image formation. Isaksen et al.13mod-eled depth of field effects using dynamically reparameter-ized light fields. Although we are also convolving imageswith blur filters that vary with depth, our filters encode theeffects of the entire optical system, not just depth of field.Furthermore, since our input consists of two-dimensionalimages, we do not have the luxury of a ray distributionbuffer. Consequently, we handle occlusion in an ad hoc man-ner.There is a significant and somewhat untapped potential forresearch that addresses the role of the human visual systemin computer graphics. One of the earliest contributions, Up-still’s Ph.D. dissertation35, considered the problem of view-ing synthetic images on a CRT and derived post-processingtechniques for improved display. Spencer et al.32investi-gated image-based techniques of adding simple ocular andcamera effects such as glare, bloom, and lenticular halo.Bolin and Meyer2used a perceptually-based sampling al-gorithm to monitor images as they are being rendered forartifacts that require a change in rendering technique. Tum-blin and Rushmeier34, Chiu et al.5, Ferweda et al.8, Ward etal.36, and Pattanaik et al.24studied the problem of mappingradiance values to the tiny fixed range supported by displaydevices. They have described a variety of tone reproductionoperators, from entirely ad hoc to perceptually based. Meyerand Greenberg22presented a color space defined by the fun-damental spectral sensitivity functions of the human visualsystem. They used this color space to modify a full color


Introducing Vision-Realistic Rendering

Download Introducing Vision-Realistic Rendering
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Introducing Vision-Realistic Rendering and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Introducing Vision-Realistic Rendering 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?