Generative art

Lecture 27

Dr. Mine Çetinkaya-Rundel

Duke University
STA 313 - Spring 2023

Warm up

Announcements

• Peer eval 2 is due today at 5pm.

• HW 06 is due Thursday at noon.

• Course + TA evaluations are due on Thursday.

• Don’t

Projects presentations

Show up at the beginning (or, better yet, 10 mins before) the time period you’re assigned to:

Any questions / things I can clarify about the projects or anything else we have covered so far?

Awards

Best thank you card

Best ugly plot

Best team name

Marvel Cinematic Tidyverse

Generative art

Setup

# load packages
library(tidyverse)
library(scico)

# set theme for ggplot2
ggplot2::theme_set(ggplot2::theme_minimal(base_size = 14))

# set width of code output
options(width = 65)

# set figure parameters for knitr
knitr::opts_chunk$set(
  fig.width = 7, # 7" width
  fig.asp = 0.618, # the golden ratio
  fig.retina = 3, # dpi multiplier for displaying HTML output on retina
  fig.align = "center", # center align figures
  dpi = 300 # higher dpi, sharper image
)

Genesis 338

by Thomas Lin Pedersen

More from Thomas Lin Pedersen

Flametree

by Danielle Navarro

More from Danielle Navarro

Permutations

by Georgios Karamanis

More from Georgios Karamanis

Abstractions

by Antonio Sánchez Chinchón

More from Antonio Sánchez Chinchón

Generative art: what

One overly simple but useful definition is that generative art is art programmed using a computer that intentionally introduces randomness as part of its creation process.

Jason Bailey

• There is randomness to the art, introduced by the computer (the code)
• The artist has control over the randomness, as contradictory as that may sound

Generative art: why

• Why people create generative art pieces?
  • Artistic expression
  • Leveraging computers to generate randomness
  • …

• Why are we learning about generative art?
  • A different look at data visualization: not for the meaning in the data, but for the visual itself
  • Great way to practice programming, particularly if you’re into creating art
  • Opportunity to practice problem solving skills, particularly if you’re sketching first, then creating or reproducing an existing piece

Generative art in use: Book covers

Just one of many examples of generative art as a practical solution by the New York Public Library:

To fill in missing book covers in an ebook-borrowing and reading app

New York Public Library, Generative eBook Covers. September 3, 2014.

I can plot myself flowers

Let’s make a circle

# make a circle of points
t <- seq(0, 2 * pi, length.out = 50)
x <- sin(t)
y <- cos(t)
df <- tibble(t, x, y)

# plot a circle of points
ggplot(df, aes(x, y)) +
  geom_point() +
  coord_equal()

The Golden Angle

The golden angle is the angle subtended by the smaller (red) arc when two arcs that make up a circle are in the golden ratio.

\[ \pi(3 − \sqrt{5}) \]

The golden angle is the angle separating successive florets on a sunflower.

Wikipedia. Golden angle.

Petals with the golden angle

# set golden angle
angle <- pi * (3 - sqrt(5))

# set number of points
points <- 50

# make data frame
df <- tibble(
  i = 1:points,
  t = (1:points) * angle,
  x = sin(t),
  y = cos(t)
)

# plot points in a spiral
ggplot(df, aes(x = x * t, y = y * t)) +
  geom_path(color = "gray") +
  geom_text(aes(label = i)) +
  coord_equal()

Paths to points

# set golden angle
angle <- pi * (3 - sqrt(5))

# set number of points
points <- 50

# make data frame
df <- tibble(
  i = 1:points,
  t = (1:points) * angle,
  x = sin(t),
  y = cos(t)
)

# plot points in a spiral
ggplot(df, aes(x = x * t, y = y * t)) +
  geom_point() +
  coord_equal()

Without the background

# set golden angle
angle <- pi * (3 - sqrt(5))

# set number of points
points <- 50

# make data frame
df <- tibble(
  i = 1:points,
  t = (1:points) * angle,
  x = sin(t),
  y = cos(t)
)

# plot points in a spiral
ggplot(df, aes(x = x * t, y = y * t)) +
  geom_point() +
  coord_equal() +
  theme_void()

More points

# set golden angle
angle <- pi * (3 - sqrt(5))

# set number of points
points <- 100

# make data frame
df <- tibble(
  i = 1:points,
  t = (1:points) * angle,
  x = sin(t),
  y = cos(t)
)

# plot points in a spiral
ggplot(df, aes(x = x * t, y = y * t)) +
  geom_point() +
  coord_equal() +
  theme_void()

Adjust points

# set golden angle
angle <- pi * (3 - sqrt(5))

# set number of points
points <- 100

# make data frame
df <- tibble(
  i = 1:points,
  t = (1:points) * angle + 20,
  x = sin(t),
  y = cos(t)
)

# plot points in a spiral
ggplot(df, aes(x = x * t, y = y * t)) +
  geom_point() +
  coord_equal() +
  theme_void()

Formalize a system

Write a function: build_art()

build_art <- function() {

  # set golden angle
  angle <- pi * (3 - sqrt(5))

  # set number of points
  points <- 100

  # make data frame
  tibble(
    i = 1:points,
    t = (1:points) * angle + 20,
    x = sin(t),
    y = cos(t)
  )
}

Add arguments to build_art()

build_art <- function(points, angle, adjustment) {
  tibble(
    i = 1:points,
    t = (1:points) * angle + adjustment,
    x = sin(t),
    y = cos(t)
  )
}

build_art(
  angle = pi * (3 - sqrt(5)),
  points = 100,
  adjustment = 20
)

# A tibble: 100 × 4
       i     t       x       y
   <int> <dbl>   <dbl>   <dbl>
 1     1  22.4 -0.398  -0.918 
 2     2  24.8 -0.327   0.945 
 3     3  27.2  0.879  -0.476 
 4     4  29.6 -0.970  -0.243 
 5     5  32.0  0.551   0.834 
 6     6  34.4  0.157  -0.988 
 7     7  36.8 -0.783   0.622 
 8     8  39.2  0.998   0.0701
 9     9  41.6 -0.688  -0.726 
10    10  44.0  0.0173  1.00  
# ℹ 90 more rows

Use the function

build_art(
  angle = pi * (3 - sqrt(5)),
  points = 100,
  adjustment = 20
) |>
  ggplot(aes(x = x * t, y = y * t)) +
  geom_point() +
  coord_equal() +
  theme_void()

Change parameters

build_art(
  angle = 3,
  points = 500,
  adjustment = 0
) |>
  ggplot(aes(x = x * t, y = y * t)) +
  geom_point() +
  coord_equal() +
  theme_void()

Next steps…

• Add random noise
• Add more customization options
  • Size
  • Color
  • Shape
  • …

Livecoding

Go to ae-22 and let’s make some flowers!

Wrap up

Acknowledgements

• Unpredictable paintings by Danielle Navarro

• Drawing Flowers with R and ggplot2 by Amanda Rowe

• Generative Art and R by the Generative Arts Collective

• Getting Started with Generative Art in R

• Why Love Generative Art? by Jason Bailey

Learn/do more

• R packages:

  • aRtsy by Koen Derks
  • generativeart by Katharina Brunner
  • flametree and jasmines bu Danielle Navarro
  • ambient and particles by Thomas Lin Pedersen

• More aRtists: rtistry art gallery by Ijeamaka Anyene

• A whole course on Generative Art by Danielle Navarro: https://art-from-code.netlify.app