Fabrications, or
How to Lie with Computer Vision

University of Kansas | 14 September 2013
Digital Humanities Forum Keynote

Jentery Sayers | U. of Victoria | @jenterysayers

–– Use your spacebar or arrow keys to navigate this slidedeck. ––
Slidedeck revised 2-4 October 2013 for publication online.
Fork this presentation.

image made using Skanect, c/o The Maker Lab

Today

Some example computer vision projects

The question of computer vision and culture

Some intersections with digital humanities

Two "Z-Axis" projects from The Maker Lab

Computer Vision

the automated description and reconstruction
of the physical world through algorithms

(image c/o Instructables)

Some Examples

CAPTCHAs

Completely Automated Public Turing test
to tell Computers and Humans Apart

(image c/o Pavel Simakov;
CAPTCHA c/o Carnegie Mellon)

Face Detection

(video by Adam Harvey)

Photogrammetry

Stitching together component parts

(image made w/ 123D Catch, c/o The Maker Lab)

Crowdsourced
Sculpture

Assembling custom parts made
across geographical distance

(image c/o PrintToPeer + Jeff de Boer)

Surveillance Forensics

extract trajectories and snapshots from video

(image c/o Computer Vision Online)

Unmanned AVs

point-and-click waypoints and scripted missions

(image c/o RCGroups; see also DIY Drones)

Marvin Minsky and Gerald Jay Sussman

"linking a camera to a computer and getting the camera to describe what it saw"

(ca. 1966, at MIT | see Szeliski 2010)

Computer Vision:
Cultural Perspectives

What's At Stake? Politically? Aesthetically?

Sensor-Vernacular

the grain of seeing, the grain of computation

(for commentary + image, see
Jones 2011 and Berry 2013)

The New Aesthetic

machine vision supplants human vision

(c/o Bridle 2011; also see Jones 2013)

The Women of ENIAC

"nameless disappearing computer operators"

(1940s, Penn + Aberdeen | see Chun 2011)

Racial Profiling and Biometrics

"falsify the idea that certain surveillance technologies and their application
are always neutral"

(see Browne 2010)

OpenCV

"make it easy to experiment with
the most common computer vision tools"

(see Borenstein 2013)

Learning from
Existing DH Projects

speculative computing + expressive models

Speculative Computing

"place the hermeneutic inside a visual and algorithmic system"

(see Drucker and Nowviskie 2004)

Non-Consumptive Reading

"Make the computer model itself
an expressive object."

(see Sample 2013)

Z-Axis in the MLab

two example Maker Lab projects
engaging computer vision

(image care of Jon Olaf Johnson)

Warped Modernisms

z-axis attempts to express time spent reading;
maps warped using CV + 3D sculpture techniques

(sparked by Alex Christie, Arthur Hain,
and Katie Tanigawa 2013)

images and research c/o
Alex Christie + Katie Tanigawa

Stereoscope Kit

uses CV to conduct a history of the senses
with an awareness of the mechanism

(in collaboration with Devon Elliott,
Shaun Macpherson, Katie McQueston,
and William J. Turkel)

images and research c/o Devon Elliott

images and research c/o Devon Elliott

Humanities on
the Z-Axis

depth | speculation | mediation | embodiment

(image care of Arthur Hain)

Thank you,
Maker Lab team

Special thanks to Nina Belojevic, Alex Christie, Devon Elliott, Arthur Hain, Jon Johnson,
Shaun Macpherson, Katie McQueston, Katie Tanigawa, and Zaqir Virani

Also to the Social Sciences and Humanities
Research Council and the Canadian Foundation
for Innovation

Thank You, #dhforum2013

Special thanks to Arienne Dwyer, Brian Rosenblum, Germaine Halegoua, Whitney Trettien, Colin Allen, Alex Gil, and the IDRH.

Keep in touch.

@jenterysayers | jentery@uvic.ca | maker.uvic.ca