Interactive segmentation: a scalable superpixel-based method

Abstract

Open Archive Toulouse Archive OuverteOATAO is an open access repository that collects the work of Toulouse researchers and makes it freely available over the web where possible Abstract. This paper addresses the problem of interactive multiclass segmentation of images. We propose a fast and efficient new interactive segmentation method called superpixel α fusion (SαF). From a few strokes drawn by a user over an image, this method extracts relevant semantic objects. To get a fast calculation and an accurate segmentation, SαF uses superpixel oversegmentation and support vector machine classification. We compare SαF with competing algorithms by evaluating its performances on reference benchmarks. We also suggest four new datasets to evaluate the scalability of interactive segmentation methods, using images from some thousand to several million pixels. We conclude with two applications of SαF.

(Retracted) Multimedia music image production combining HMM and song feature tags

“…The relatively free rhythm processing can be felt from the interpretation of the four performance versions. 26 The characteristics of the four performance versions are shown in Fig. 4.…”

“…Four performers (Barenboim performance version, Pires performance version, Polini performance version, and Rubinstein performance version) have performed local rhythm expansion and contraction here. The relatively free rhythm processing can be felt from the interpretation of the four performance versions 26 . The characteristics of the four performance versions are shown in Fig.…”

TissueWand, a Rapid Histopathology Annotation Tool

“…[ 6 7 ] In region-based methods, users typically click-drag to paint categories to signify pixels that lie within the segments. [ 8 9 ] Typically, as a result of initial inputs, the resulting full-image segmentation is visualized as a shaded overlay or as polygons delimiting categories.…”

“…[ 11 ] To speed up processing at interaction time, a precomputed oversegmentation can be used with algorithms applied to these larger segments rather than on raw pixels. [ 9 ]…”