Study 1C: Machine Vision

"But over the last decade or so, something dramatic has happened. Visual culture has changed form. It has become detached from human eyes and has largely become invisible. Human visual culture has become a special case of vision, an exception to the rule. The overwhelming majority of images are now made by machines for other machines, with humans rarely in the loop. The advent of machine-to-machine seeing has been barely noticed at large, and poorly understood by those of us who’ve begun to notice the tectonic shift invisibly taking place before our very eyes."

Trevor Paglen – Invisible Images (Your Pictures Are Looking at You)

Overview

Today the bulk of images created are for machine consumption – many of which are never meant to be seen by humans. From Automatic License Plate Readers (ALPR) and optical character recognition (OCR) used to photograph and document license plates, to vision and motion tracking software to aid advertisers in public places, machines are often times creating images for other machines in order to collect information about humans.

And while we create imagery for each other – the photos on our social media feeds, the graphics we make as designers, and the like – they are largely facilitated by machines who. For instance, when an image is uploaded to Facebook algorithms work to identify your face and tag you. Similarly, iOS, Google photos, and other software all process your photos for varying outcomes – from recognizing patterns and automatically making Gifs, to auto-creating albums for individuals.

In preparation of saving a photo, machines create images as they process our likeness and detect their surroundings for effects like faceswap and autofocus. it is arguable such processeses are constantly making and discarding images every ‘frame’ that they process.

To address the topic, you’ll create a test graphic for the technology of your choice. Considering the social and technological implications of machine vision and your technology in particular.

Learning outcomes

  • Identify the limits of facial recognition technologies
  • Develop graphics that respond to machine as well as human perception
  • Consider the role of machine vision in visual culture today

Calendar

Week 9
T – Assign Project + In Class Exercise
TR – Present image collections for ‘Shirley Cards’

Week 10
Spring Break

Week 11
T – Rough draft Shirley Cards

Week 12
T – Final Crit with screenshots of outcomes

Reading

Prompt

Step 1

In groups of two, choose a camera or software to investigate. This can be your phone, a digital camera, or an application such as Snapchat, Snow, etc. Select a technology / software you have access to, and are interested in working with.

Then run a series of tests with your technology to test its technological boundaries. For instance, see which areas of interest/focus emerge in tests of images, the boundaries of face detection technology, how filters react to people of different races, etc.

Due Tuesday Mar 13th

Step 2

Based on your research, collect 2 proposals sets of images to use for your own ‘Shirley Card’. This proposal should take into account the technology you are investigating, as well as social concerns. Collect your images and lay them out on a tabloid page using design software. Present your proposals digitally.

Due Thursday Mar 15th

Step 3

Choose a proposal and design and produce an 11x17 in. printed ‘Shirley Card’ for the technology of choice. While traditional test cards are used for printing, and color calibration, your card will test the boundaries of your machine vision program.

Present the ‘Shirley Card’ along with a series of screenshots or digital images of your technology reacting to your test card as a slideshow, website, or video loop.

Due Tuesday April 2nd

Requirements

  • 2 proposal image collections presented digitally
  • 11x17 in printed "Shirley Card"
  • Screen-based documentation of your "Shirley Card" in use

References

back to home