[AG-USERS] TODAY | Real-Time Illumination Models and Adaptive Frameless Ray Tracing

[Cindy Sievers]

LANL is hosting a technical talk today, Tuesday March 7 (today) at 1:00 - 
3:00pm Mountain time.  All sites are welcome.  We will test 1/2 hour before 
the talk, and vnc information will be distributed then.


>***  CCS-1 Seminar Series
>***  http://public.ds.lanl.gov/ccs1-seminar
>
>TITLE:
>Bridging the Gap: Real-Time Illumination Models and Adaptive
>Frameless Ray Tracing
>
>SPEAKER:
>David Luebke, University of Virgina,
>http://www.cs.virginia.edu/~luebke, see also bio below
>
>WHEN:
>Tuesday, March 7, 1pm Mountain time
>
>WHERE:
>Access Grid room, TA-3, building 200

Virtual Venue: Platinum


>ABSTRACT:
>Rendering research tends to fall into two camps. Realistic rendering
>seeks to synthesize photorealistic images offline, taking seconds,
>minutes, or even hours per image, while real-time rendering targets
>interactivity, producing imagines in milliseconds but sacrificing
>realism by the use of simpler, more hardware-friendly models for
>geometry, illumination, camera lenses, etc.  The overarching goal of
>my research group has been to bridge this gap between realistic and
>real-time graphics. I will give a preview of two current projects, one
>that works to increase realism in interactive rendering and the other
>that helps accelerate realistic rendering techniques into the
>interactive realm.  First, our work on real-time illumination models
>builds on precomputed radiance transfer to enable some phenomena
>previously unaddressed in interactive graphics: rendering of glossy
>objects with interreflections, indirect lighting, and sharp shadows,
>and rendering of translucent objects with both single and multiple
>subsurface scattering. The key innovations behind these contributions
>involve new applications of non-linear wavelet approximation and
>matrix factorization to the reflectance and scattering equations, as
>well as efficient algorithms for GPU implementation. Second, our work
>on adaptive frameless ray tracing (AFRT) exploits the ability of ray
>tracing, traditionally an offline realistic rendering technique, to
>selectively sample the image plane. AFRT replaces traditional gridded,
>framed sampling with a stochastic & spatio-temporally adaptive
>approach. This leads to the interesting question, "How should we
>display an image in which some pixels are older than others?". We use
>a novel reconstruction that responds to temporal image gradients:
>Where the displayed scene is static, spatial color change dominates
>and older samples are given significant weight in reconstruction,
>resulting in sharper and eventually antialiased images. Where the
>scene is dynamic, more recent samples are emphasized, resulting in
>less sharp but fully up-to-date images.  We have shown that AFRT can
>produce images of comparable accuracy to traditional rendering
>techniques, using an order of magnitude fewer samples.
>
>BIO:
>David Luebke is an Assistant Professor in the Department of Computer
>Science at the University of Virginia. He earned his Ph.D. in Computer
>Science at the University of North Carolina in 1998 under Frederick
>P. Brooks, Jr., and earned his Bachelors degree in Chemistry at the
>Colorado College. Professor Luebke's principal research interest is
>interactive computer graphics, particularly the problem of acquiring
>and rendering very complex real-world scenes at interactive
>rates. Specific projects include polygonal level of detail (LOD),
>temperature-aware graphics architecture, scientific computation on
>graphics hardware, advanced reflectance and illumination models for
>real-time rendering, and image-based acquisition of real-world
>environments.  Another project was the Virtual Monticello museum
>exhibit, which ran for over 3 months and helped attract over 110,000
>visitors as a centerpiece of the major exhibition "Jefferson's America
>and Napoleon's France" at the New Orleans Museum of Art. Luebke is
>also lead author of the book "Level of Detail for 3D Graphics".
>
>


============================================
Cindy Sievers           Los Alamos National Laboratory
sievers at lanl.gov        Group CCS-1 MS B287
tel:505.665.6602        Advanced Computing
fax:505.665.4939        Los Alamos, NM 87544
============================================