An interactive system for structure-based ASCII art creation

Non-Photorealistic Rendering (NPR), whose aim is to create artistic style images, is one of the important research topics in computer graphics. One example of NPR is an art form called ASCII art which represents pictures using character strings. ASCII art is commonly used in media that cannot display images or mainly use text, such as e-mail and bulletin board system. ASCII art can be categorized into two styles, tone-based style and structure-based style. The structurebased style can represent content by using less number of ASCII characters compared to the tone-based style. However, in general, it takes beginners a long time to create structurebased ASCII art. As such, an automatic method that creates structure-based ASCII art is required to reduce the tasks of ASCII art creation. In this paper, we propose an interactive system which creates structure-based ASCII art. We provide four matching metrics for converting an image into an ASCII art.Based on experimental results, we found that the suitable matching metrics to produce visually pleasant ASCII art images depend on the type of the input images. Since our system can produce ASCII art images in a few seconds, users can create several ASCII art images using different matching metrics, then select the one they like the most.