Creativity Enhancement of Painterly Rendering using a Suggestive Interface
Non-photorealistic rendering (NPR) can use various parameters and techniques to automatically generate a wide range of images with different painterly appearances. However, a real painting is the result of a creative process, and traditionally, many artists have developed their own media, styles, and techniques. Fortunately, the computer can also be used in different ways, for instance, as a tool to support and enhance creativity. We propose a system to interactively generate a painterly image through a suggestive interface. After the artist intuitively selects a region, our system immediately offers different suggestions of NPR modifications. The multidimensional NPR parameter space is transformed in a more perceptual space by optimization, which is validated by a user study. The suggested images generated from sampled parametric variations are displayed according to our perceptual distance. Through the same interface, our system allows for refinement or exploration without any need to understand the various parameters involved in generating NPR effects. The interest of our suggestive interface is supported by another user study, where participants felt that it stimulated their creativity as they interactively and iteratively created painterly images by choosing suggested images for the selected regions. Another experiment showed that different viewers found the resulting images more creative than their initial automatically generated images.