Co-segmentation via visualization

Kamranian Z, Tombari F, Nilchi ARN, Monadjemi A, Navab N (2018)

Publication Type: Journal article

Publication year: 2018

Journal

Journal of Visual Communication and Image Representation Elsevier

Book Volume: 55

Pages Range: 201-214

DOI: 10.1016/j.jvcir.2018.05.014

Abstract

This paper addresses the co-segmentation problem using feature visualization for CNNs. Visualization is exploited as an auxiliary information to discriminate salient image regions (dubbed as “heat-regions”) from non-salient ones. Region occlusion sensitivity is proposed for feature visualization. The co-segmentation problem is formulated via a convex quadratic optimization which is initialized by the heat-regions. The information obtained through the visualization is considered as an extra energy term in the cost function. The results of the visualization demonstrate that there exist some heat-regions which are not productive in the co-segmentation. To detect helpful regions among them, an adaptive strategy in the form of an iterative algorithm is proposed according to the consistency among all images. Comparison experiments conducted on two benchmark datasets, iCoseg and MSRC, illustrate the superior performance of the proposed approach over state-of-the-art algorithms.

Involved external institutions

University of Isfahan (UI) / Esfahan University / دانشگاه اصفهان IR Iran, Islamic Republic of (IR) Technische Universität München (TUM) DE Germany (DE)

How to cite

APA:

Kamranian, Z., Tombari, F., Nilchi, A.R.N., Monadjemi, A., & Navab, N. (2018). Co-segmentation via visualization. Journal of Visual Communication and Image Representation, 55, 201-214. https://doi.org/10.1016/j.jvcir.2018.05.014

MLA:

Kamranian, Zahra, et al. "Co-segmentation via visualization." Journal of Visual Communication and Image Representation 55 (2018): 201-214.

BibTeX: Download