The relatively recent rise in popularity of paint by numbers has stemmed from its ability to be used as a creative or therapeutic activity. Paint by numbers kits are straightforward and easy to use, but they provide a unique way for people to express themselves artistically. Despite the benefits, creating these kits is highly time consuming and requires the skills of a designer, advanced digital equipment, and high levels of effort. Properly segmenting an image, mapping the correct colors, and making the kit usable for novices is the primary issue with this approach. All of the techniques available are heavily reliant on a person’s knowledge, making it nearly impossible to adjust the scale. This paper describes an approach to automating the creation of paint by numbers canvases using modern techniques in image processing. The approach relies on the combination of vector-based segmentation, color clustering algorithms, and optimized workflows with a means to convert arbitrary images into paint by number templates by segmentation and clustering of colors. The approach described above allows for improved efficiency by greatly decreasing the amount of manual effort needed.

The growing demand for creative and recreational activities has led to the development of an AI-powered platform that automates paint-by-numbers canvas generation. By integrating image segmentation, color extraction (using K-means clustering), and vectorization, the system transforms user-uploaded images into customizable paint-by-number templates with precise color palettes. Leveraging OpenAI’s DALL-E for image generation, the platform simplifies a traditionally labor-intensive process while maintaining high-quality outputs. Usability tests confirm high user satisfaction due to its intuitive interface and high-resolution results. Beyond artistic applications, the tool also offers therapeutic benefits by promoting mindfulness and stress reduction, bridging the gap between technology and creative expression for both hobbyists and businesses.