Saliency-Based Artistic Abstraction With Deep Learning and Regression Trees

Abstract

ion in art often reflects human perception-areas of an artwork that hold the observer's gaze longest will generally be more detailed, while peripheral areas are abstracted, just as they are mentally abstracted by humans' physiological visual process. The authors' artistic abstraction tool, Salience Stylize, uses Deep Learning to predict the areas in an image that the observer's gaze will be drawn to, which informs the system about which areas to keep the most detail in and which to abstract most. The planar abstraction is done by a Random Forest Regressor, splitting the image into large planes and adding more detailed planes as it progresses, just as an artist starts with tonally limited masses and iterates to add fine details, then completed with our stroke engine. The authors evaluated the aesthetic appeal and effectiveness of the detail placement in the artwork produced by Salience Stylize through two user studies with 30 subjects.

Assessing Heterogeneity in Students’ Visual Judgment: Model-Based Partitioning of Image Rankings

“…Even though the intended ranking for abstract images was moderately low (between 29% and 42%), the worth parameters did not reflect the presence of outlying responses between student groups, i.e., there was no large systematic difference in ranking order among students. Different sets of stimuli, e.g., computer generated art that controls for salience ( Furnham and Rao, 2002 ; Shakeri et al, 2017 ) with a focus on a single dimensions of visual abstraction, such as composition or color ( Markovic, 2010 ) could lead to higher variability in perceived judgment.…”