From global image annotation to interactive object segmentation

• Published in: Multimedia tools and applications Vol. 70; no. 1; pp. 475 - 493
• Main Authors: Giró-i-Nieto, Xavier, Martos, Manuel, Mohedano, Eva, Pont-Tuset, Jordi
• Format: Journal Article Publication
• Language: English
• Published: Boston Springer US 01.05.2014; Springer Nature B.V
• Subjects: Algorithms; Analysis; Annotations; Computer Communication Networks; Computer Science; Crowdsourcing; Data processing; Data Structures and Information Theory; Enginyeria de la telecomunicació; Image analysis; Image processing systems; Imatge; Multimedia computer applications; Multimedia Information Systems; Ontology; Pattern recognition; Processament; Processament del senyal; Semantics; Software; Special Purpose and Application-Based Systems; Studies; Tècniques digital; Àrees temàtiques de la UPC; Hierarchical; Segmentation; Interaction; Multiscale; Annotation
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-013-1374-3

This paper presents a graphical environment for the annotation of still images that works both at the global and local scales. At the global scale, each image can be tagged with positive, negative and neutral labels referred to a semantic class from an ontology. These annotations can be used to train and evaluate an image classifier. A finer annotation at a local scale is also available for interactive segmentation of objects. This process is formulated as a selection of regions from a precomputed hierarchical partition called Binary Partition Tree. Three different semi-supervised methods have been presented and evaluated: bounding boxes, scribbles and hierarchical navigation. The implemented Java source code is published under a free software license.