Generative Art

library(ggplot2)
library(dplyr)

Attaching package: 'dplyr'

The following objects are masked from 'package:stats':

    filter, lag

The following objects are masked from 'package:base':

    intersect, setdiff, setequal, union

library(ggforce)
library(ambient)
library(MetBrewer)
library(metR)      
library(scales)  
library(colorspace)

iris_art <- function(data = iris) {
  
  dat <- data %>%
    #transform iris data into something that is plottable polarly
    mutate(
      x0 = (Sepal.Length - min(Sepal.Length)) / (max(Sepal.Length) - min(Sepal.Length)),
      y0 = (Sepal.Width - min(Sepal.Width)) / (max(Sepal.Width) - min(Sepal.Width)),
      x1 = x0 + (Petal.Length - min(Petal.Length)) / (max(Petal.Length) - min(Petal.Length)) * 0.3,
      y1 = y0 + (Petal.Width - min(Petal.Width)) / (max(Petal.Width) - min(Petal.Width)) * 0.3,
      size = Sepal.Length / max(Sepal.Length)
    )
  # green palette
  paquin_pal <- met.brewer("Veronese")[1:3]  
  
  ggplot(dat, aes(x = x0, y = y0, xend = x1, yend = y1, colour = Species, size = size)) +
    geom_segment(lineend = "round", show.legend = FALSE) +
    coord_polar() +
    scale_colour_manual(values = paquin_pal) +
    scale_size(range = c(0.5, 3)) +
    theme_void()+
    theme(plot.background = element_rect(fill = "beige", color = NA))
}

iris_art()

Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
ℹ Please use `linewidth` instead.

Title: Twist of Iris

At first glance, Twist of Iris feels like a a flower frozen mid-spin, caught in the act of opening. But stare a little longer, and the lines start to echo something else, an iris of an eye.

This piece takes the iris dataset, with each segment built from sepal and petal measurements, “twists” into movement and direction. The use of the a green palette lends the composition a natural, organic feel, like petals dancing in sunlight, pollen drifting in the wind, or a person’s eye. It plays with both meanings of “iris”(a flower and an eye ).

# generate each individual polygon
generate_polygon <- function(center_x, center_y, radius, n_points) {
  angles <- seq(0, 2 * pi, length.out = n_points + 1)[-1]
  angles <- jitter(angles, amount = 0.2)
  data.frame(
    x = center_x + radius * cos(angles),
    y = center_y + radius * sin(angles)
  )
}

palette <- met.brewer("Ingres", n = 9, type = "continuous")
# blue gold palette

set.seed(42)
n_polygons <- 300
polygon_data <- bind_rows(lapply(1:n_polygons, function(i) {
  center_x <- runif(1, -10, 10)
  center_y <- runif(1, -10, 10)
  radius <- runif(1, 0.2, 0.4) # shape size
  n_points <- sample(3:7, 1)
  base_color <- sample(palette, 1)
  
  polygon <- generate_polygon(center_x, center_y, radius, n_points)
  polygon$group <- i
  polygon$fill <- base_color
  polygon
}))

ggplot(polygon_data, aes(x = x, y = y, group = group, fill = fill)) +
  # added an edge line to the shapes
  geom_shape(alpha = 0.8, radius = unit(0.1, 'cm'), color = "#736545", size = 0.1) +
  coord_equal() +
  theme_void() +
  scale_fill_identity() +
  theme(plot.background = element_rect(fill = "black", color = NA))

Title: Isolated Pebbles

Description:

Isolated Pebbles presents scattered stones, each rendered in earthy tones with a stark black void-like background. The colors of the stones suggests they are grounded and along with the irregularity of shape reminds the viewer of pebbles one might find on a riverside. These anchor each shape firmly against the deep black background, suggesting they rest quietly on unseen soil.

While some pebbles appear clustered, almost friends, a second look reveals that none truly touch. Even shapes that share similar colors and forms(as with traits that people have in common) remain separated by subtle distances, and are not groped together, suggesting an inherent solitude. This conveys how closeness does not always equate to connection, and similarity does not guarantee friendship. This also could suggest that there are many people like you out there you might yet to have find them though!