Channel Decomposition into Painting Actions
This is an incremental method for generating artistic styles in image processing, potentially useful for artists or creative applications.
The paper tackles the problem of decomposing convolutional layers into painting actions by simulating human painter behaviors, such as hand movement and stroke style, and uses Mask R-CNN for object detection to plan painting order, resulting in extensions to artistic styles based on parameters.
This work presents a method to decompose a convolutional layer of the deep neural network into painting actions. To behave like the human painter, these actions are driven by the cost simulating the hand movement, the paint color change, the stroke shape and the stroking style. To help planning, the Mask R-CNN is applied to detect the object areas and decide the painting order. The proposed painting system introduces a variety of extensions in artistic styles, based on the chosen parameters. Further experiments are performed to evaluate the channel penetration and the channel sensitivity on the strokes.