Mid-Level concept learning with visual contextual ontologies and probabilistic inference for image annotation

Authors:
Yuee Liu;Jinglan Zhang;Dian Tjondronegoro;Shlomo Geva;Zhengrong Li
Affiliations:
Faculty of Science and Technology, Queensland Univerisity of Technology, Brisbane, Australia;Faculty of Science and Technology, Queensland Univerisity of Technology, Brisbane, Australia;Faculty of Science and Technology, Queensland Univerisity of Technology, Brisbane, Australia;Faculty of Science and Technology, Queensland Univerisity of Technology, Brisbane, Australia;Faculty of Science and Technology, Queensland Univerisity of Technology, Brisbane, Australia
Venue:
MMM'10 Proceedings of the 16th international conference on Advances in Multimedia Modeling
Year:
2010

Citing 5
Cited 0

A Critical View of Context

International Journal of Computer Vision
Image-Segmentation Evaluation From the Perspective of Salient Object Extraction

CVPR '06 Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1
Incorporating concept ontology to enable probabilistic concept reasoning for multi-level image annotation

MIR '06 Proceedings of the 8th ACM international workshop on Multimedia information retrieval
LabelMe: A Database and Web-Based Tool for Image Annotation

International Journal of Computer Vision
Learning Spatial Context: Using Stuff to Find Things

ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

To date, automatic recognition of semantic information such as salient objects and mid-level concepts from images is a challenging task. Since real-world objects tend to exist in a context within their environment, the computer vision researchers have increasingly incorporated contextual information for improving object recognition. In this paper, we present a method to build a visual contextual ontology from salient objects descriptions for image annotation. The ontologies include not only partOf/kindOf relations, but also spatial and co-occurrence relations. A two-step image annotation algorithm is also proposed based on ontology relations and probabilistic inference. Different from most of the existing work, we exploit how to combine representation of ontology, contextual knowledge and probabilistic inference. The experiments in the LabelMe dataset show that image annotation results are improved using contextual knowledge.