paint selection siggraph09.pdf

Paint Selection Jiangyu Liu? University of Science and Technology of China Jian Sun? Microsoft Research Asia Heung-Yeung Shum? Microsoft Corporation Figure 1: Left three: the user makes a selection by painting the object of interest with a brush (black-white circle) on a 24.5 megapixel image. Instant feedback (selection boundary or image effect) can be provided to the user during mouse dragging. Rightmost: composition and effect (sepia tone). Note that the blue scribbles are invisible to the user. They are drawn in the paper for illustration only. Abstract. In this paper, we present Paint Selection, a progressive painting-based tool for local selection in images. Paint Selection facilitates users to progressively make a selection by roughly paint- ing the object of interest using a brush. More importantly, Paint Selection is efficient enough that instant feedback can be provided to users as they drag the mouse. We demonstrate that high quality selections can be quickly and effectively “painted” on a variety of multi-megapixel images. Keywords: image segmentation, user interface 1 Introduction Selections or layers are one of the most powerful representations in image editing. Many applications, from object cut-and-paste to local color/tone adjustments, require a local selection. Recently, graph-cut-based approaches [Boykov and Jolly 2001; Li et al. 2004; Rother et al. 2004] have greatly simplified this tedious and time- consuming job. Furthermore, efficient optimization [Boykov and Kolmogorov 2001] also enables users to get instant feedback, which is critical for interactive image editing. ?This work was done when Jiangyu Liu was an intern at Microsoft Re- search Asia. Jiangyu Liu is with MOE-Microsoft Key Laboratory of MCC, USTC. Email: eeyu@ ?Email: {jiansun, hshum}@ Gradually, however, these global optimization based approaches become incapable of providing instant feedback on multi- megapixel (10–25Mp) images produced by today’s digital cameras, because the compl