Sisley the abstract painter

Abstract

We present an interactive abstract painting system named Sisley. Sisley works upon the psychological principle [Berlyne 1971] that abstract arts are often characterized by their greater perceptual ambiguities than photographs, which tend to invoke moderate mental efforts of the audience for interpretation, accompanied with subtle aesthetic pleasures. Given an input photograph, Sisley decomposes it into a hierarchy/tree of its constituent image components (e.g., regions, objects of different categories) with interactive guidance from the user, then automatically generates corresponding abstract painting images, with increased ambiguities of both the scene and individual objects at desired levels. Sisley consists of three major working parts: (1) an interactive image parser executing the tasks of segmentation, labeling, and hierarchical organization, (2) a painterly rendering engine with abstract operators for transferring the image appearance, and (3) a numerical ambiguity computation and control module of servomechanism. With the help of Sisley, even an amateur user can create abstract paintings from photographs easily in minutes. We have evaluated the rendering results of Sisley using human experiments, and verified that they have similar abstract effects to original abstract paintings by artists.

Stroke Based Painterly Rendering

“…Zhao and Zhu [27] propose to boost color saturation proportional to the underlying importance map computed for the source image. The SBR process of point sampling color, and generating marks of greater size than one pixel, may be regarding as an integration or low-pass filtering process.…”

“…The link between the scene and the marks is the style. Other higher level aspects of the scene such as geometry and composition are typically not manipulated, though there has been some progress in this area [4,27].…”

Evaluation of emotional response to non-photorealistic images

“…Nonetheless, researchers in NPR often seek to improve the aesthetic qualities of their synthetic images. Zhao and Zhu [2010] use aesthetic arguments (about the appeal of abstract paintings) to motivate their algorithm, and perform a user study measuring the ease with which subjects could recognize objects in their synthetic semi-abstract paintings. Duke et al [2003] investigated the emotional impact of a few types of images, though in an unsystematic way and without using images that were rated for affect: the coverage of emotional space was sparse and the range of styles covered was small.…”

“…The method of Kang et al [2007] was used to produce line art: among automatic image-based methods, the results of this algorithm are unsurpassed. For the painterly algorithms, we faced a difficult decision because of the wide variety of methods and the different styles of output they generate; ultimately we decided to use two algorithms, the classic Haeberli "Paint by Numbers" [1990] and the recent "Sisley" [Zhao and Zhu 2010]. We consider Haeberli's method to produce more representational images (closer to the original) while the output from Sisley is more abstract.…”

“…To determine how the perceived emotional content of the image is affected by the different algorithms, we conducted a study investigating the emotional response of 42 users to five well-known NPR algorithms (Haeberli's interactive painting [Haeberli 1990]; the photo abstraction of Orzan et al [Orzan et al 2007]; the abstract painting of Zhao et al [Zhao and Zhu 2010]; Secord's weighted Voronoi stippling [Secord 2002]; and the line drawing techique of Kang et al [Kang et al 2007]), two blurring techniques that were used to reduce the image information in a systematic manner (uniform blur and salience-based blur), and the original image. Emotional response was measured using the three-dimensional model of emotion, which consists of arousal (level of activation), valence (positive versus negative), and dominance (level of control).…”