Generating Pseudo-3D Painting Based on Visual Saliency and Composition Rules


Zhibin Zheng, Yan Zhang, Zhengxing Sun

Short paper,accepted to EuroGraphics (Proc. EG 2013)



This paper proposes a cropping method based on visual saliency and composition rules to get pseudo-3D painting froma 2D given image. On the basis of segmenting the given image into foreground and background, we utilize visualsaliency to determine the highlighted region and combine existing composition rules to automatically crop theimage using PSO optimization algorithm. Additionally, we add the border and pseudo-perspective transformationto enhance stereoscopic effect of our results. Experimental results demonstrate our method could turn a reference image into a pseudo-3D painting effectively, bringing a stereoscopic feeling to observers.


Overview


An overview of our approach. With an input image(a), we first segment the image into background and foregroud(b), and then detect the saliency area of the image(c), then we fit the foreground object as a ellipse, and finally crop the image into pseudo-3D painting(d).



Comparison


We compare our results with [RTMS12].The left image is input image, middle is our result, right is the result from [RTMS12].





Other Results


Our pseudo results by adding border.




PAPER: Generating Pseudo-3D Painting Based on Visual Saliency and Composition Rules

[RTMS12].RITSCHEL T., TEMPLIN K., MYSZKOWSKI K., SEIDELH.-P.: Virtual passepartouts. In Proceedings of the Symposiumon Non-Photorealistic Animation and Rendering (Airela-Ville, Switzerland, Switzerland, 2012), NPAR ’12, pp. 57–63. http://dl.acm.org/citation.cfm?id=2330147.2330160. 1, 4