Abstract
In this paper we discuss certain theoretical properties of the algorithm selection approach to the problem of semantic segmentation in computer vision. We show that an algorithm's score depends on final task. Thus to properly evaluate an algorithm and to determine its suitability, precise score value obtained on well formulated tasks can be used only. When an algorithm suitability is well known, the algorithm can be efficiently used for a task by applying it in the most favorable environmental conditions determined during the evaluation. However, high quality algorithm selection is possible only if each algorithm suitability is well known because only then the algorithm selection result can improve the best possible result given by a single algorithm. The task dependent evaluation is demonstrated on segmentation and object recognition. Additionally, we also discuss the importance of high level symbolic knowledge in the selection process. The importance of this symbolic hypothesis is demonstrated on a set of learning experiments with both a Bayesian Network and SVM. We show that task dependent evaluation is required to allow efficient algorithm selection. Also by studying symbolic preference of algorithms for semantic segmentation we show that algorithm selection accuracy can be improved by 10 to 15%.
| Original language | English |
|---|---|
| Title of host publication | International Conference on Information and Digital Technologies, IDT 2015 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 223-229 |
| Number of pages | 7 |
| ISBN (Print) | 9781467371858 |
| DOIs | |
| Publication status | Published - Aug 25 2015 |
| Event | International Conference on Information and Digital Technologies, IDT 2015 - Zilina, Slovakia Duration: Jul 7 2015 → Jul 9 2015 |
Other
| Other | International Conference on Information and Digital Technologies, IDT 2015 |
|---|---|
| Country | Slovakia |
| City | Zilina |
| Period | 7/7/15 → 7/9/15 |
Fingerprint
Keywords
- Algorithm Selection
- Machine Learning
- Semantic Segmentation
ASJC Scopus subject areas
- Artificial Intelligence
- Computer Science Applications
- Information Systems
Cite this
Evaluation of algorithm selection approach for semantic segmentation based on high-level information feedback. / Lukac, Martin; Abdiyeva, Kamila; Kameyama, Michitaka.
International Conference on Information and Digital Technologies, IDT 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 223-229 7222974.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Evaluation of algorithm selection approach for semantic segmentation based on high-level information feedback
AU - Lukac, Martin
AU - Abdiyeva, Kamila
AU - Kameyama, Michitaka
PY - 2015/8/25
Y1 - 2015/8/25
N2 - In this paper we discuss certain theoretical properties of the algorithm selection approach to the problem of semantic segmentation in computer vision. We show that an algorithm's score depends on final task. Thus to properly evaluate an algorithm and to determine its suitability, precise score value obtained on well formulated tasks can be used only. When an algorithm suitability is well known, the algorithm can be efficiently used for a task by applying it in the most favorable environmental conditions determined during the evaluation. However, high quality algorithm selection is possible only if each algorithm suitability is well known because only then the algorithm selection result can improve the best possible result given by a single algorithm. The task dependent evaluation is demonstrated on segmentation and object recognition. Additionally, we also discuss the importance of high level symbolic knowledge in the selection process. The importance of this symbolic hypothesis is demonstrated on a set of learning experiments with both a Bayesian Network and SVM. We show that task dependent evaluation is required to allow efficient algorithm selection. Also by studying symbolic preference of algorithms for semantic segmentation we show that algorithm selection accuracy can be improved by 10 to 15%.
AB - In this paper we discuss certain theoretical properties of the algorithm selection approach to the problem of semantic segmentation in computer vision. We show that an algorithm's score depends on final task. Thus to properly evaluate an algorithm and to determine its suitability, precise score value obtained on well formulated tasks can be used only. When an algorithm suitability is well known, the algorithm can be efficiently used for a task by applying it in the most favorable environmental conditions determined during the evaluation. However, high quality algorithm selection is possible only if each algorithm suitability is well known because only then the algorithm selection result can improve the best possible result given by a single algorithm. The task dependent evaluation is demonstrated on segmentation and object recognition. Additionally, we also discuss the importance of high level symbolic knowledge in the selection process. The importance of this symbolic hypothesis is demonstrated on a set of learning experiments with both a Bayesian Network and SVM. We show that task dependent evaluation is required to allow efficient algorithm selection. Also by studying symbolic preference of algorithms for semantic segmentation we show that algorithm selection accuracy can be improved by 10 to 15%.
KW - Algorithm Selection
KW - Machine Learning
KW - Semantic Segmentation
UR - http://www.scopus.com/inward/record.url?scp=84954349181&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84954349181&partnerID=8YFLogxK
U2 - 10.1109/DT.2015.7222974
DO - 10.1109/DT.2015.7222974
M3 - Conference contribution
AN - SCOPUS:84954349181
SN - 9781467371858
SP - 223
EP - 229
BT - International Conference on Information and Digital Technologies, IDT 2015
PB - Institute of Electrical and Electronics Engineers Inc.
ER -