Weakly supervised deep semantic segmentation using CNN and ELM with semantic candidate regions

Xinying Xu, Guiqing Li, Gang Xie, Jinchang Ren, Xinlin Xie

Research output: Contribution to journalArticle

7 Downloads (Pure)

Abstract

The task of semantic segmentation is to obtain strong pixel-level annotations for each pixel in the image. For fully supervised semantic segmentation, the task is achieved by a segmentation model trained using pixel-level annotations. However, the pixel-level annotation process is very expensive and time-consuming. To reduce the cost, the paper proposes a semantic candidate regions trained extreme learning machine (ELM) method with image-level labels to achieve pixel-level labels mapping. In this work, the paper casts the pixel mapping problem into a candidate region semantic inference problem. Specifically, after segmenting each image into a set of superpixels, superpixels are automatically combined to achieve segmentation of candidate region according to the number of image-level labels. Semantic inference of candidate regions is realized based on the relationship and neighborhood rough set associated with semantic labels. Finally, the paper trains the ELM using the candidate regions of the inferred labels to classify the test candidate regions. The experiment is verified on the MSRC dataset and PASCAL VOC 2012, which are popularly used in semantic segmentation. The experimental results show that the proposed method outperforms several state-of-the-art approaches for deep semantic segmentation.

Original languageEnglish
Article number9180391
Number of pages12
JournalComplexity
Volume2019
DOIs
Publication statusPublished - 14 Mar 2019

Fingerprint

Semantics
Learning systems
Pixels
Labels
Machine Learning
Segmentation
Machine learning
Volatile organic compounds
Costs and Cost Analysis
Annotation
Costs

Keywords

  • semantic segmentation
  • image semantic segmentation
  • image detection

Cite this

Xu, Xinying ; Li, Guiqing ; Xie, Gang ; Ren, Jinchang ; Xie, Xinlin. / Weakly supervised deep semantic segmentation using CNN and ELM with semantic candidate regions. In: Complexity. 2019 ; Vol. 2019.
@article{09aa9fbe90754a379f596ea1c85db230,
title = "Weakly supervised deep semantic segmentation using CNN and ELM with semantic candidate regions",
abstract = "The task of semantic segmentation is to obtain strong pixel-level annotations for each pixel in the image. For fully supervised semantic segmentation, the task is achieved by a segmentation model trained using pixel-level annotations. However, the pixel-level annotation process is very expensive and time-consuming. To reduce the cost, the paper proposes a semantic candidate regions trained extreme learning machine (ELM) method with image-level labels to achieve pixel-level labels mapping. In this work, the paper casts the pixel mapping problem into a candidate region semantic inference problem. Specifically, after segmenting each image into a set of superpixels, superpixels are automatically combined to achieve segmentation of candidate region according to the number of image-level labels. Semantic inference of candidate regions is realized based on the relationship and neighborhood rough set associated with semantic labels. Finally, the paper trains the ELM using the candidate regions of the inferred labels to classify the test candidate regions. The experiment is verified on the MSRC dataset and PASCAL VOC 2012, which are popularly used in semantic segmentation. The experimental results show that the proposed method outperforms several state-of-the-art approaches for deep semantic segmentation.",
keywords = "semantic segmentation, image semantic segmentation, image detection",
author = "Xinying Xu and Guiqing Li and Gang Xie and Jinchang Ren and Xinlin Xie",
year = "2019",
month = "3",
day = "14",
doi = "10.1155/2019/9180391",
language = "English",
volume = "2019",
journal = "Emergence",
issn = "1521-3250",

}

Weakly supervised deep semantic segmentation using CNN and ELM with semantic candidate regions. / Xu, Xinying; Li, Guiqing; Xie, Gang; Ren, Jinchang; Xie, Xinlin.

In: Complexity, Vol. 2019, 9180391, 14.03.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Weakly supervised deep semantic segmentation using CNN and ELM with semantic candidate regions

AU - Xu, Xinying

AU - Li, Guiqing

AU - Xie, Gang

AU - Ren, Jinchang

AU - Xie, Xinlin

PY - 2019/3/14

Y1 - 2019/3/14

N2 - The task of semantic segmentation is to obtain strong pixel-level annotations for each pixel in the image. For fully supervised semantic segmentation, the task is achieved by a segmentation model trained using pixel-level annotations. However, the pixel-level annotation process is very expensive and time-consuming. To reduce the cost, the paper proposes a semantic candidate regions trained extreme learning machine (ELM) method with image-level labels to achieve pixel-level labels mapping. In this work, the paper casts the pixel mapping problem into a candidate region semantic inference problem. Specifically, after segmenting each image into a set of superpixels, superpixels are automatically combined to achieve segmentation of candidate region according to the number of image-level labels. Semantic inference of candidate regions is realized based on the relationship and neighborhood rough set associated with semantic labels. Finally, the paper trains the ELM using the candidate regions of the inferred labels to classify the test candidate regions. The experiment is verified on the MSRC dataset and PASCAL VOC 2012, which are popularly used in semantic segmentation. The experimental results show that the proposed method outperforms several state-of-the-art approaches for deep semantic segmentation.

AB - The task of semantic segmentation is to obtain strong pixel-level annotations for each pixel in the image. For fully supervised semantic segmentation, the task is achieved by a segmentation model trained using pixel-level annotations. However, the pixel-level annotation process is very expensive and time-consuming. To reduce the cost, the paper proposes a semantic candidate regions trained extreme learning machine (ELM) method with image-level labels to achieve pixel-level labels mapping. In this work, the paper casts the pixel mapping problem into a candidate region semantic inference problem. Specifically, after segmenting each image into a set of superpixels, superpixels are automatically combined to achieve segmentation of candidate region according to the number of image-level labels. Semantic inference of candidate regions is realized based on the relationship and neighborhood rough set associated with semantic labels. Finally, the paper trains the ELM using the candidate regions of the inferred labels to classify the test candidate regions. The experiment is verified on the MSRC dataset and PASCAL VOC 2012, which are popularly used in semantic segmentation. The experimental results show that the proposed method outperforms several state-of-the-art approaches for deep semantic segmentation.

KW - semantic segmentation

KW - image semantic segmentation

KW - image detection

UR - http://www.scopus.com/inward/record.url?scp=85063528277&partnerID=8YFLogxK

U2 - 10.1155/2019/9180391

DO - 10.1155/2019/9180391

M3 - Article

VL - 2019

JO - Emergence

JF - Emergence

SN - 1521-3250

M1 - 9180391

ER -