. .

The Evolutionary Poster Composer

The Evolutionary Poster Composer is a set of generative experiments that employs a semi-automatic Evolutionary Computation approach for generating, evolving, and evaluating typographical posters. The main motivation behind these experiments is to develop a generative process that replicates the workflow of traditional letterpress print-houses. Letterpress is a printing technique that became popular on the follow-up of the Industrial Revolution because it allowed a cheaper, easier, and faster printing of commercial posters for mass communication. However, the design process of these posters was slightly different from the typical contemporary poster design process. At the time, letterpress designers created posters while trying to fill a matrix, often in collaboration with the client. The design decisions of these designers were very pragmatic: extensive sentences were composed in condensed typefaces, and short sentences were composed in extended typefaces. Also, the most important parts of the content were emphasised using bigger typefaces.

 

Evolutionary Poster Composer at  exhibition
Figure 1

Evolutionary Poster Composer at xCoAx 2018 exhibition (c arte c, Madrid, Spain, 2018).

 

In these experiments, we employed Evolutionary Computation approaches because they can encapsulate the basic set of tasks and activities that often graphic designers perform. This way, these approaches may be extremally useful tools for graphic designers, especially for visual and conceptual exploration, in the most exploratory stages of their projects. Briefly, the system employs a Genetic Algorithm to interactively arranges a certain text, divided into text boxes, in order to fulfil a predefined canvas divided into a grid. This text may be inputted by the user or gathered automatically from multiple web sources. After, a population of candidate solutions (or posters) is evolved using variation operators (i.e. recombination and mutation). The system is also able to evaluate the outputs generated. Also, the system’s users may guide the evolutionary process according to their visual and conceptual preferences. Currently, we have developed two distinct semi-automatic experiments.

 

We are interested in replicating this technique using computational approaches mainly with three main research and conceptual aims. First, the letterpress design process is by itself an algorithmic process (i.e. the designers’ crafts posters thought the use of the same «recipe») and participative (i.e. the client takes an active role in the definition of the final output). This way, this approach has the potential to allow the study of how computational design techniques may expand the tools and automate some process in Graphic Design practice, creating novel ways to communicate with people. Second, the letterpress became popular due to technological and social advances promoted by the Industrial Revolution. Currently, we believe that the democratisation of digital technologies and Artificial Intelligence will have a similar, or greater, impact in the future practice of Graphic Design, changing their production process. Finally, the letterpress poster are artefacts for mass communication and, therefore, they had an influential role in the definition of the contemporary social nature of posters. However, in these experiments, we explore posters as media for communicating personal and ephemeral information, a relatively unexplored subject that goes against this nature of letterpress posters, that promotes the discussion about the future of poster itself and about its value as media for artistic exploration.

 


 

Example generation of a poster using content a the poster designed by Atelier Populaire “Usines, Universités, Union” (1968) and the typeface Titling Gothic FB, David Berlow (Font Bureau, 2005)

Figure 2

 

Approach 1

In this first approach, the system generates outputs employing a Genetic Algorithm that composes the content over a canvas divided into a modular grid. The generation process begins with the random generation of a population of candidate solution (or posters). After, these individuals are evolved throughout recombination and mutation on the most promising individuals. The fitness of each output is calculated employing a hardwired fitness assignment scheme. This way, the merit of each generated poster is defined by weighted arithmetic mean of three parameters: (i) composition, i.e. how the poster composition fulfils the whole space on canvas; (ii) visual design, i.e. how content is possible to read on the poster; and (iii) typography, i.e. how much the typographic decision of system please the typographic preferences of users. By empirical exploration, we defined the following weights for each one of the parameters: (i) 33% for the composition; (ii) 47% for the visual design; and (iii) 20% for the typographic decisions based on the user typographical preferences. The users may guide the evolutionary process communicating their typographical preferences to the system, defining the horizontal motion and weight that please they. Figure 3 display some results more results may be found here and videos of the generative process may be found here.

 

Figure 3

Snapshot of system’s interface during evolutionary process. Poster content retrieved from the poster “Amerika is Devouring its Children” designed by Jay Belloi (1970).

 

Figure 4

Evolutionary Poster Composer in exhibition at Processing Community Day @fbaup (2019, Porto, Portugal) Photography by Pedro Amado

 

Also, we developed the experiment Digital Poster (De)Composer, when we used this evolutionary approach and, after, we misshapen the outputs using a basic pixel deformation algorithm.

 

Figure 5

Digital Poster (De)Composer in featured at Mostra Jovens Criadores 2019 exhibition (Nova BSE, Carcavelos, Oeiras, January 2020)

 

 

Approach 2

In the second approach, the system generates outputs also employing a Genetic Algorithm that composes the content over a one-row column grid. Such as in the approach 1, the generation process begins with the random generation of a population of candidate solution (or posters) and, after, these individuals are evolved throughout recombination and mutation on the most promising individuals. In this approach the users may communicate their preferences to the system by defining and interactively updating, several settings parameters in a dedicated interface (see figure 6). This version of the system is available online here.

 

Figure 6

Snapshot of approach 2 system’s interface during evolutionary process. Poster content retrieved from book The Arcades Project of Space of Walter Benjamin (1974)

 

The fitness of each output is calculated employing a hardwired fitness assignment scheme that evaluates three objectives: (i) legibility (i.e. how much content it is possible to read on the poster); (ii) aesthetics (i.e. how much the design of the poster satisfies a set of aesthetics measures for typographic poster design); and (iii) semantics (i.e. how much the visual features of the poster convey the semantic meaning of its content). The overall merit of each poster is defined by the weighted arithmetic mean of these three objectives. The weight of each objective on the mean is defined in the interface. Example results may be found here and a demonstration video of the system may be found here.

 
 

Related

The Posters’ Factory
Evolutionary Poster Composer at xCoAx 2018

 
 

Exibhitions and Installations

Poster Factory. In: Interactive Video Installations Exibhitions of Golden bee 14 (un)real Global Biennale of Graphic Design
Experiments in the Development of Typographical Posters. In: Exhibition of the 6th Conference on Computation, Communication, Aesthetics & X (XCoAx 2018). Centro arte computulense, Madrid, Spain.
Evolutionary Poster Composer. In: Processing Community Day Porto 2019. Faculty of Fine Arts of University of Porto, Portugal.
Digital Poster (De)Composer. In: Mostra Nacional Jovens Criadores 2018. Nova BSE, Oeiras, Portugal.
The Poster’s Factory. In: Y, Designing Portugal – Projects by National Design Schools. Porto Design Bienalle 2019. 22th November—23 February 2020, Galeria Municipal Matosinhos, Portugal.

 

Publications

  • S. Rebelo, C. M. Fonseca, J. Bicker, and P. Machado, “Evolutionary Experiments in the Development of Typographical Posters,” in 6th Conference on Computation, Communication, Aesthetics & X, Madrid, Spain (xCoAx 2018), 2018, pp. 65-75.

  • S. Rebelo, J. Bicker, and P. Machado, “Evolutionary Experiments in Typesetting of Letterpress-Inspired Posters,” in Proceedings of the Eleventh International Conference on Computational Creativity, September 7–11, 2020, 2020.