Visual Computing for Computer Architects: Imaging, Computing, and Human Perception

Tutorial Co-located with ASPLOS 2023

9:00 AM — 12:00 PM, March 26, 2023, Junior AB

University of Rochester
New York University
teaser

Why This Tutorial?

Over the past two decades, computer architecture research has moved from general-purpose computing toward domain-specific accelerators. The field is on the cusp of a new revolution — human-centric architectures — driven by emerging visual computing platforms such as Augmented Reality, Virtual Reality, and autonomous machines that all intimately interact with humans.

The goal of this tutorial is to set up computer systems and architecture researchers for the era of human-centered visual computing, both to build efficient systems guided by human cognition and to augment human cognition through computing technologies. Critically, this requires us to bridge the conventional computing domain with imaging and human perception, the two fundamental components that connect computing with humans.

To that end, this tutorial will 1) teach the fundamental principles behind visual computing, including imaging (e.g., optics, image sensors), computing (on visual content), and human perception and cognition, and 2) discuss potential research opportunities for the systems and architecture community. The tutorial is not tied to a specific benchmark or hardware platform. Instead, we focus on teaching the fundamental principles so that participants can walk away being able to design new algorithms, building new applications, and engineering systems and hardware for those algorithms and applications.


Agenda

Yuhao Zhu (9:00 AM -- 9:10 AM)
Yuhao Zhu (9:10 AM -- 10:20 AM)
  • Fundamental principles of imaging
  • Camera optics
  • Modern CMOS image sensor architecture
  • Camera image-signal processing and hardware
  • Imaging-computing interface
  • Computational image sensors
Coffer Break
10:20 AM -- 10:40 AM
Qi Sun (10:40 AM -- 12:00 AM)
  • Basic facts of human vision: sptial/temporal/color vision
  • Computational models of human Vision
  • Computational holographic displays
  • Gaze-contingent rendering techniques
  • Neural rendering