Accuracy improvement of functional attribute recognition by dense CRF considering object shape

Masaki Iizuka, Shuichi Akizuki, Manabu Hashimoto

Research output: Contribution to journalArticle

Abstract

In this paper, we propose a method to recognize functional attributes of everyday objects for vision system of partner robots. On the related research, there is a method to optimize recognition result with dense (fully connected) conditional random field (CRF), which uses the estimation result of functional attribute for each pixel. However, since this method is optimized from RGB data, it is not able to sufficiently consider the shape of object, which has a relationship with the function attribute. In the proposed method, the recognition accuracy of functional attributes is improved by considering the object shape with the dense CRF describing the three-dimensional positional relationship. As a result of the experiment, the recognition rate of the proposed method is 77.0%, which is 3.8% higher than the related method. In addition, we confirmed that the processing speed is high as a side effect by reducing processing cost by oversegmentation of input data and using high-speed identification by Random Forests. The mean processing speed per an object was 109 ms in the proposed method.

Original languageEnglish
JournalElectronics and Communications in Japan
DOIs
Publication statusAccepted/In press - 2019 Jan 1

Fingerprint

Conditional Random Fields
Attribute
Processing
high speed
Pixels
robots
Robots
pixels
costs
Random Forest
Vision System
Object
Costs
Experiments
High Speed
Pixel
Robot
Optimise
Three-dimensional
Experiment

Keywords

  • affordance
  • dense CRF
  • functional attribute
  • partner robot

ASJC Scopus subject areas

  • Signal Processing
  • Physics and Astronomy(all)
  • Computer Networks and Communications
  • Electrical and Electronic Engineering
  • Applied Mathematics

Cite this

Accuracy improvement of functional attribute recognition by dense CRF considering object shape. / Iizuka, Masaki; Akizuki, Shuichi; Hashimoto, Manabu.

In: Electronics and Communications in Japan, 01.01.2019.

Research output: Contribution to journalArticle

@article{4cbf0e6f7f7f46a699cf8f1eda38c29a,
title = "Accuracy improvement of functional attribute recognition by dense CRF considering object shape",
abstract = "In this paper, we propose a method to recognize functional attributes of everyday objects for vision system of partner robots. On the related research, there is a method to optimize recognition result with dense (fully connected) conditional random field (CRF), which uses the estimation result of functional attribute for each pixel. However, since this method is optimized from RGB data, it is not able to sufficiently consider the shape of object, which has a relationship with the function attribute. In the proposed method, the recognition accuracy of functional attributes is improved by considering the object shape with the dense CRF describing the three-dimensional positional relationship. As a result of the experiment, the recognition rate of the proposed method is 77.0{\%}, which is 3.8{\%} higher than the related method. In addition, we confirmed that the processing speed is high as a side effect by reducing processing cost by oversegmentation of input data and using high-speed identification by Random Forests. The mean processing speed per an object was 109 ms in the proposed method.",
keywords = "affordance, dense CRF, functional attribute, partner robot",
author = "Masaki Iizuka and Shuichi Akizuki and Manabu Hashimoto",
year = "2019",
month = "1",
day = "1",
doi = "10.1002/ecj.12151",
language = "English",
journal = "Electronics and Communications in Japan",
issn = "1942-9533",
publisher = "Scripta Technica",

}

TY - JOUR

T1 - Accuracy improvement of functional attribute recognition by dense CRF considering object shape

AU - Iizuka, Masaki

AU - Akizuki, Shuichi

AU - Hashimoto, Manabu

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In this paper, we propose a method to recognize functional attributes of everyday objects for vision system of partner robots. On the related research, there is a method to optimize recognition result with dense (fully connected) conditional random field (CRF), which uses the estimation result of functional attribute for each pixel. However, since this method is optimized from RGB data, it is not able to sufficiently consider the shape of object, which has a relationship with the function attribute. In the proposed method, the recognition accuracy of functional attributes is improved by considering the object shape with the dense CRF describing the three-dimensional positional relationship. As a result of the experiment, the recognition rate of the proposed method is 77.0%, which is 3.8% higher than the related method. In addition, we confirmed that the processing speed is high as a side effect by reducing processing cost by oversegmentation of input data and using high-speed identification by Random Forests. The mean processing speed per an object was 109 ms in the proposed method.

AB - In this paper, we propose a method to recognize functional attributes of everyday objects for vision system of partner robots. On the related research, there is a method to optimize recognition result with dense (fully connected) conditional random field (CRF), which uses the estimation result of functional attribute for each pixel. However, since this method is optimized from RGB data, it is not able to sufficiently consider the shape of object, which has a relationship with the function attribute. In the proposed method, the recognition accuracy of functional attributes is improved by considering the object shape with the dense CRF describing the three-dimensional positional relationship. As a result of the experiment, the recognition rate of the proposed method is 77.0%, which is 3.8% higher than the related method. In addition, we confirmed that the processing speed is high as a side effect by reducing processing cost by oversegmentation of input data and using high-speed identification by Random Forests. The mean processing speed per an object was 109 ms in the proposed method.

KW - affordance

KW - dense CRF

KW - functional attribute

KW - partner robot

UR - http://www.scopus.com/inward/record.url?scp=85059691631&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059691631&partnerID=8YFLogxK

U2 - 10.1002/ecj.12151

DO - 10.1002/ecj.12151

M3 - Article

JO - Electronics and Communications in Japan

JF - Electronics and Communications in Japan

SN - 1942-9533

ER -