Kayvon photo
Assistant Professor of Computer Science
Stanford University
Contact Info
Gates Building 366
I architect high-performance visual computing systems that enable immersive and intelligent visual computing applications. In pursuit of these goals, my recent research efforts can be categorized into two main themes:
The Visual Data Analysis Engine. The next generation of visual computing applications will require efficient analysis and mining of large repositories of visual data (images, videos, RGBD). But scaling visual data analysis to operate on collections the size of all public photos and videos on Facebook, all security video cameras in a major city, or petabytes of images in an astronomy sky survey, presents supercomputing-scale storage and computation challenges. Very few programmers have the capability to operate efficiently at these scales, inhibiting the field's exploration of advanced data-driven visual computing applications. To meet this challenge, we are developing a distributed computing platform -- combining ideas from high-performance image processing languages, data analytics, and database functionality -- that facilitates the development of applications that query, analyze and mine image and video collections at scale.
The Graphics Engine Compiler. We are developing new programming abstractions and compiler frameworks for graphics engines of the future -- engines that will deliver rich virtual worlds to platforms ranging from beefy GPUs powering VR headsets to energy-limited operation on mobile devices. A key aspect of this project is enabling rapid exploration of the complex space of optimization decisions present in these systems. We're tackling the problem using new shading languages that separate key optimization decisions from high-level program specification and via automatic shader optimization techniques that make approximations (like automatic code simplification) to achieve desirable performance-quality trade-offs. We seek to make it possible to rapidly generate highly efficient graphics systems specialized to application content or target machine architecture.
I'm always looking for great students that wish to work on these topics, or bring their own great ideas.
I recently gave an Arch 2030 workshop keynote on how visual computing applications will drive architectual innovation in the year 2030. Here is the talk on Youtube, and an updated version of the slides is here.
Before moving to Stanford, I taught the following courses at CMU.
Here are a few tips on how to give clear research talks (or class project talks).
I created this talk, Do Grades Matter, to challenge students at CMU to think bigger than just striving to get good grades in a bunch of hard classes.
Former students:
Shader Components: Modular and High Performance Shader Development
Yong He, Tim Foley, Teguh Hofstee, Haomin Long, Kayvon Fatahalian
Automatically Scheduling Halide Image Processing Pipelines
Ravi Teja Mullapudi, Andrew Adams, Dillon Sharlet, Jonathan Ragan-Kelley, Kayvon Fatahalian
A System for Rapid Exploration of Shader Optimization Choices
Yong He, Tim Foley, Kayvon Fatahalian
LED Street Light Research Project Part II: New Findings
Stephen Quick, Donald Carter, Kayvon Fatahalian, Cynthia Limauro
CMU Technical Report, Summer 2016
The Rise of Mobile Visual Computing Systems
Kayvon Fatahalian
IEEE Pervasive Computing, April/June 2016
Automatically Splitting a Two-Stage Lambda Calculus
Nicolas Feltman, Carlo Anguili, Umut A. Acar, Kayvon Fatahalian
European Symposium on Programming (ESOP) 2016
KrishnaCam: Using a Longitudinal, Single-Person, Egocentric Dataset for Scene Understanding Tasks
Krishna Kumar Singh, Kayvon Fatahalian, Alexei Efros
WACV 2016
A System for Rapid, Automatic Shader Level-of-Detail
Yong He, Tim Foley, Natalya Tatarchuk, Kayvon Fatahalian
SIGGRAPH Asia 2015
Aggregate G-Buffer Anti-Aliasing
Cyril Crassin, Morgan McGuire, Kayvon Fatahalian, Aaron Lefohn
I3D 2015
(An updated and extended version of the paper appears in TVCG 2016.)
Extending the Graphics Pipeline with Adaptive, Multi-Rate Shading
Yong He, Yan Gu, Kayvon Fatahalian
Self-Refining Games using Player Analytics
Matt Stanton, Ben Humberston, Brandon Kase, James O'Brien, Kayvon Fatahalian, Adrien Treuille
Near-exhaustive Precomputation of Secondary Cloth Effects
Doyub Kim, Woojong Koh, Rahul Narain, Kayvon Fatahalian, Adrien Treuille, James O'Brien
Efficient BVH Construction via Approximate Agglomerative Clustering
Yan Gu, Yong He, Kayvon Fatahalian, Guy Blelloch
High Performance Graphics 2013
SRDH: Specializing BVH Construction and Traversal Order Using Representative Shadow Ray Sets
Nicolas Feltman, Minjae Lee, Kayvon Fatahalian
High Performance Graphics 2012
Evolving the Real-Time Graphics Pipeline for Micropolygon Rendering
Kayvon Fatahalian, Stanford University Ph.D. Dissertation, 2011
Reducing Shading on GPUs using Quad-Fragment Merging
Kayvon Fatahalian, Solomon Boulos, James Hegarty, Kurt Akeley, William R. Mark, Henry Moreton, Pat Hanrahan
Space-Time Hierarchical Occlusion Culling for Micropolygon Rendering with Motion Blur
Solomon Boulos, Edward Luong, Kayvon Fatahalian, Henry Moreton, Pat Hanrahan
High Performance Graphics 2010
Hardware Implementation of Micropolygon Rasterization with Motion and Defocus Blur
John S. Brunhaver, Kayvon Fatahalian, Pat Hanrahan
High Performance Graphics 2010
A Lazy Object-Space Shading Architecture With Decoupled Sampling
Christopher A. Burns, Kayvon Fatahalian, William R. Mark
High Performance Graphics 2010
DiagSplit: Parallel, Crack-Free, Adaptive Tessellation for Micropolygon Rendering
Matthew Fisher, Kayvon Fatahalian, Solomon Boulos, Kurt Akeley, William R. Mark, Pat Hanrahan
SIGGRAPH Asia 2009
Data-Parallel Rasterization of Micropolygons with Defocus and Motion Blur
Kayvon Fatahalian, Edward Luong, Solomon Boulos, Kurt Akeley, William R. Mark, Pat Hanrahan
High Performance Graphics 2009
GRAMPS: A Programming Model for Graphics Pipelines
Jeremy Sugerman, Kayvon Fatahalian, Solomon Boulos, Kurt Akeley, Pat Hanrahan
Transactions on Graphics (TOG) January 2009
A Closer Look at GPUs
Kayvon Fatahalian and Mike Houston
Communications of the ACM. Vol. 51, No. 10 (October 2008)
(also published as "GPUs: A Closer Look": ACM Queue. March/April. 2008)
A Portable Runtime Interface for Multi-level Memory Hierarchies
Mike Houston, Ji Young Park, Manman Ren, Timothy J. Knight, Kayvon Fatahalian, Alex Aiken, William J. Dally, Pat Hanrahan
PPOPP 2008
Compilation for Explicitly Managed Memory Hierarchies
Timothy J. Knight, Ji Young Park, Manman Ren, Mike Houston, Mattan Erez, Kayvon Fatahalian, Alex Aiken, William J. Dally, Pat Hanrahan
PPOPP 2007
Sequoia: Programming the Memory Hierarchy
Kayvon Fatahalian, Timothy J. Knight, Mike Houston, Mattan Erez, Daniel R Horn, Larkhoon Leem, Ji Young Park, Manman Ren, Alex Aiken, William J. Dally, Pat Hanrahan
Supercomputing 2006
Understanding the Efficiency of GPU Algorithms for Matrix-Matrix Multiplication
Kayvon Fatahalian, Jeremy Sugerman, Pat Hanrahan
Graphics Hardware 2004
Brook for GPUs: Stream Computing on Graphics Hardware
Ian Buck, Tim Foley, Daniel Horn, Jeremy Sugerman, Kayvon Fatahalian, Mike Houston, Pat Hanrahan
Precomputing Interactive Dynamic Deformable Scenes
Doug L. James and Kayvon Fatahalian
Real-Time Global Illumination of Deformable Objects
Undergraduate Senior Research Thesis (Carnegie Mellon University). 2003.
Advised by Doug James
Self-Refining Interactive Games (graphics with 100's of machines and a lot of latency)
How do we build platforms that take graphics applications from one user on a single GPU to 10,000 machines and one million users in the cloud? Even though computer graphics has always been at the vanguard of parallel computing, there has been little success using modern cloud-based computing resources to improve interactive experiences. In this project we asked the question, how could we leverage the massive storage and batch processing capabilities of the cloud to generate new forms of interactive worlds -- and we took a "precompute everything" approach to doing so. Since one cannot precompute everything about an complex interactive world, the challenge is to determine what is most important to precompute, so these parts can be presented to the user with the highest-quality graphics. We find that by recording statistics of users playing a game, we can build a model of user behavior, and then concentrate large-scale, cloud-based precomputation of graphics and physics around the states that users are most likely to encounter. The result is a self-refining game whose dynamics improve with play, ultimately providing realistically rendered, rich fluid dynamics in real time on a mobile device. For more detail, see our work applied these ideas to cloth simulation and fluid simulation.
A Real-Time Micropolygon Rendering Pipeline (evolving the GPU pipeline for tiny triangles)
GPUs will soon have the compute horsepower to render scenes containing cinematic-quality surfaces in real-time. Unfortunately, if they render these subpixel polygons (micropolygons) using the same techniques as they do for large triangles today, GPUs will perform extremely inefficiently. Instead of trying to parallelize Pixar's Reyes micropolygon rendering system, we're taking a hard look at how the existing Direct3D 11 rendering pipeline, and GPU hardware implementations, must evolve to render micropolygon workloads efficiently in a high-throughput system. Changes to software interfaces, algorithms, and HW design are fair game! Slides describing what we've learned can be found in this SIGGRAPH course talk or in my dissertation: Evolving the Real-Time Graphics Pipeline for Micropolygon Rendering.
GRAMPS (a framework for heterogeneous parallel programming)
There are two ways to think about GRAMPS. Graphics folks should think of GRAMPS as a system for building custom graphics pipelines. We simply gave up on adding more and more configurable knobs to existing pipelines like OpenGL/Direct3D and instead allow the programmer to programmatically define a custom pipeline with an arbitrary number of stages connected by queues. To non-graphics folks, GRAMPS is a stream programming system that embraces heterogeneity in underlying architecture and anticipates streaming workloads that exhibit both regular and irregular (dynamic) behavior. The GRAMPS runtime dynamically schedules GRAMPS programs onto architectures containing a mixture of compute-optimized cores, generic CPU cores, and fixed-function processing units.
The Sequoia Programming Language ("Programming the Memory Hierarchy")
Sequoia is a hierarchical stream programming language that arose from the observation that expressing locality, not parallelism is the most important responsibility of parallel application programmers in scientific/numerical domains. Sequoia presents a parallel machine as an abstract hierarchy of memories and gives the programmer explicit control over data locality and communication through this hierarchy using first-class language constructs (basically, Sequoia supports nested kernels and streams of streams). Sequoia programs have run on a variety of exposed-communication architectures such as clusters, the CELL processor, GPUs, and even supercomputing clusters at Los Alamos. The best way to learn about Sequoia is to read our SC06 paper. You can also learn more at the Sequoia project page.
Brook/Merrimac (stream processing for scientific computing)
I helped out with the BrookGPU (abstracting the GPU as a stream processor for numerical computing) and Merrimac Streaming Supercomputer projects.

My work is supported by the National Science Foundation, the Heinz Foundation, and by INTEL, NVIDIA, QUALCOMM, GOOGLE, ADOBE, FACEBOOK, ACTIVISION, and APPLE.