Gotta Catch 'em AIl

Co-creative Pokémon Generation

By Christina Irakleous, Nikki Rademaker, and Yanna Smid

Overview

For the course Computational Creativity, we created a system that lets users design new Pokémon with the help of AI. The project combines creativity with generative AI tools, allowing users to pick types, body shapes, and an inspiration theme. Based on these choices, the system generates both a textual description and a visual representation of a new Pokémon.

The goal was to build a simple, creative tool that makes the Pokémon design process more accessible and interactive.

• Design Input: Users can select pokémon types, body shape, and write down inspiration.
• Pokémon Description: GPT-3.5-Turbo generates a text description of the new Pokémon.
• Pokémon Visual: DALL-E 3 creates a matching image based on the generated description.

The images below show the interface and what users are expected to do to use the tool.

Process

The following steps were completed to create this project:

• Concept Development: we quickly decided to work on a project combining AI and video games, based on our shared interest in Pokémon.
• Sketching: we designed a simple and clear interface to guide users through the input process.
• Implementation: we built the interface in Python using Tkinter and connected it to OpenAI's GPT-3.5-Turbo and DALL-E 3 APIs.
• Optimization: we refined the prompts for the AI to generate more fitting Pokémon designs.
• Evaluation: we evaluated the tool using creativity measures (SPECS) and a small Turing test.

Technical Development

• The interface was built with Python and Tkinter.
• For text Generation, Pokémon descriptions were created with OpenAI GPT-3.5-Turbo.
• Image generation was done with OpenAI DALL-E 3 based on the generated text.

Generator Outputs

The Pokémon generator successfully created convincing and creative Pokémon. Most users could not tell the AI designs apart from real ones, meaning it passed the Turing Test. According to the SPECS evaluation, the system was found creative, although some Pokémon looked slightly less realistic.

Below are a few example outputs from the Pokémon generator: