Abstract
Previous image classification approaches mostly neglect semantics, which has two major limitations. First, categories are simply treated independently while in fact they have semantic overlaps. For example, “sedan” is a specific kind of “car”. Therefore, it’s unreasonable to train a classifier to distinguish between “sedan” and “car”. Second, image feature representations used for classifying different categories are the same. However, the human perception system is believed to use different features for different objects. In this paper, we leverage semantic ontologies to solve the aforementioned problems. The authors propose an ontological random forest algorithm where the splitting of decision trees are determined by semantic relations among categories. Then hierarchical features are automatically learned by multiple-instance learning to capture visual dissimilarities at different concept levels. Their approach is tested on two image classification datasets. Experimental results demonstrate that their approach not only outperforms state-of-the-art results but also identifies semantic visual features.
| Original language | English (US) |
|---|---|
| Title of host publication | Computer Vision |
| Subtitle of host publication | Concepts, Methodologies, Tools, and Applications |
| Publisher | IGI Global |
| Pages | 784-799 |
| Number of pages | 16 |
| ISBN (Electronic) | 9781522552055 |
| ISBN (Print) | 9781522552048 |
| DOIs | |
| State | Published - Jan 1 2018 |
Fingerprint
ASJC Scopus subject areas
- Computer Science(all)
Cite this
Ontological random forests for image classification. / Xu, Ning; Wang, Jiangping; Qi, Guojun; Huang, Thomas S; Lin, Weiyao.
Computer Vision: Concepts, Methodologies, Tools, and Applications. IGI Global, 2018. p. 784-799.Research output: Chapter in Book/Report/Conference proceeding › Chapter
}
TY - CHAP
T1 - Ontological random forests for image classification
AU - Xu, Ning
AU - Wang, Jiangping
AU - Qi, Guojun
AU - Huang, Thomas S
AU - Lin, Weiyao
PY - 2018/1/1
Y1 - 2018/1/1
N2 - Previous image classification approaches mostly neglect semantics, which has two major limitations. First, categories are simply treated independently while in fact they have semantic overlaps. For example, “sedan” is a specific kind of “car”. Therefore, it’s unreasonable to train a classifier to distinguish between “sedan” and “car”. Second, image feature representations used for classifying different categories are the same. However, the human perception system is believed to use different features for different objects. In this paper, we leverage semantic ontologies to solve the aforementioned problems. The authors propose an ontological random forest algorithm where the splitting of decision trees are determined by semantic relations among categories. Then hierarchical features are automatically learned by multiple-instance learning to capture visual dissimilarities at different concept levels. Their approach is tested on two image classification datasets. Experimental results demonstrate that their approach not only outperforms state-of-the-art results but also identifies semantic visual features.
AB - Previous image classification approaches mostly neglect semantics, which has two major limitations. First, categories are simply treated independently while in fact they have semantic overlaps. For example, “sedan” is a specific kind of “car”. Therefore, it’s unreasonable to train a classifier to distinguish between “sedan” and “car”. Second, image feature representations used for classifying different categories are the same. However, the human perception system is believed to use different features for different objects. In this paper, we leverage semantic ontologies to solve the aforementioned problems. The authors propose an ontological random forest algorithm where the splitting of decision trees are determined by semantic relations among categories. Then hierarchical features are automatically learned by multiple-instance learning to capture visual dissimilarities at different concept levels. Their approach is tested on two image classification datasets. Experimental results demonstrate that their approach not only outperforms state-of-the-art results but also identifies semantic visual features.
UR - http://www.scopus.com/inward/record.url?scp=85045026777&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85045026777&partnerID=8YFLogxK
U2 - 10.4018/978-1-5225-5204-8.ch031
DO - 10.4018/978-1-5225-5204-8.ch031
M3 - Chapter
AN - SCOPUS:85045026777
SN - 9781522552048
SP - 784
EP - 799
BT - Computer Vision
PB - IGI Global
ER -