Unimore logo AImageLab


Segmentation models diversity for object proposals

Abstract: In this paper we present a segmentation proposal method which employs a box-hypotheses generation step followed by a lightweight segmentation strategy. Inspired by interactive segmentation, for each automatically placed bounding-box we compute a precise segmentation mask. We introduce diversity in segmentation strategies enhancing a generic model performance exploiting class-independent regional appearance features. Foreground probability scores are learned from groups of objects with peculiar characteristics to specialize segmentation models. We demonstrate results comparable to the state-of-the-art on PASCAL VOC 2012 and a further improvement by merging our proposals with those of a recent solution. The ability to generalize to unseen object categories is demonstrated on Microsoft COCO 2014.


Manfredi, Marco; Grana, Costantino; Cucchiara, Rita; Smeulders, Arnold W. M. "Segmentation models diversity for object proposals" COMPUTER VISION AND IMAGE UNDERSTANDING, vol. 158, pp. 40 -48 , 2017 DOI: 10.1016/j.cviu.2016.06.005

 not available

Paper download: