Convolutional Oriented Boundaries : From Image Segmentation to High-Level Tasks

We present Convolutional Oriented Boundaries (COB), which produces multiscale oriented contours and region hierarchies starting from generic image classification Convolutional Neural Networks (CNNs). COB is computationally efficient, because it requires a single CNN forward pass for multi-scale cont...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE transactions on pattern analysis and machine intelligence. - 1979. - 40(2018), 4 vom: 05. Apr., Seite 819-833
1. Verfasser:	Maninis, Kevis-Kokitsi (VerfasserIn)
Weitere Verfasser:	Pont-Tuset, Jordi, Arbelaez, Pablo, Van Gool, Luc
Format:	Online-Aufsatz
Sprache:	English
Veröffentlicht:	2018
Zugriff auf das übergeordnete Werk:	IEEE transactions on pattern analysis and machine intelligence
Schlagworte:	Journal Article Research Support, Non-U.S. Gov't