Event Location: NSH 1305
Bio: Kayvon Fatahalian is an Assistant Professor in the Computer Science Department at Carnegie Mellon University.
His research focuses on the design of programming abstractions and efficient parallel systems for computer-intensive applications such as interactive computer graphics.

Abstract: In the past fifteen years the capabilities of real-time graphics systems have increased rapidly as a
result of domain-influenced co-design of algorithms, programming interfaces, and heterogeneous,
parallel hardware (GPUs). Despite these improvements, rendering geometrically complex, film-
quality scenes at interactive rates remains beyond the capabilities of current systems. Current
GPU implementations not only require additional compute capability to handle high-resolution
surfaces represented by pixel-sized “micropolygons”, fundamental system operations such as
geometry processing, surface visibility, and shading execute inefficiently under this workload.

In this talk I will describe an evolution of the OpenGL graphics pipeline that significantly in-
creases system efficiency when rendering micropolygons. The redesign modifies rendering al-
gorithms, GPU hardware, and programming abstractions to increase parallelism and eliminate
redundant work while preserving the graphics pipeline’s simple parallel programming model and
the throughput-optimized design of GPU processing cores. This talk focuses on the challenges
of real-time graphics, but it is exciting to consider how future computational cameras and mobile
computer vision tasks will soon benefit from similar levels of system optimization.