Pixel-wise ordinal classification for salient object grading

Driven by business intelligence applications for rating attraction of products in shops, a new problem — salient object grading is studied in this paper. In computer vision, plenty of salient object detection approaches have been proposed, while most existing studies detect objects in a binary manne...

Full description

Saved in:
Bibliographic Details
Main Authors: Liu, Yanzhu, Wang, Yanan, Kong, Adams Wai Kin
Other Authors: School of Computer Science and Engineering
Format: Article
Language:English
Published: 2022
Subjects:
Online Access:https://hdl.handle.net/10356/160069
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-160069
record_format dspace
spelling sg-ntu-dr.10356-1600692022-07-12T05:45:31Z Pixel-wise ordinal classification for salient object grading Liu, Yanzhu Wang, Yanan Kong, Adams Wai Kin School of Computer Science and Engineering Engineering::Computer science and engineering Ordinal Classification Salient Object Grading Driven by business intelligence applications for rating attraction of products in shops, a new problem — salient object grading is studied in this paper. In computer vision, plenty of salient object detection approaches have been proposed, while most existing studies detect objects in a binary manner: salient or not. This paper focuses on a new problem setting that requires detecting all salient objects and categorizing them into different salient levels. Based on that, a pixel-wise ordinal classification method is proposed. It consists of a multi-resolution saliency detector which detects and segments objects, an ordinal classifier which grades pixels into different salient levels, and a binary saliency enhancer which sharpens the difference between non-saliency and all other salient levels. Two new image datasets with salient level labels are constructed. Experimental results demonstrate that, on the one hand, the proposed method provides effective salient level predictions and on the other hand, offers very comparable performance with state-of-the-art salient object detection methods in the traditional problem setting. Ministry of Education (MOE) This work is supported by Ministry of Education, Singapore (MOE2016-T2-1-042(S)). 2022-07-12T05:45:31Z 2022-07-12T05:45:31Z 2021 Journal Article Liu, Y., Wang, Y. & Kong, A. W. K. (2021). Pixel-wise ordinal classification for salient object grading. Image and Vision Computing, 106, 104086-. https://dx.doi.org/10.1016/j.imavis.2020.104086 0262-8856 https://hdl.handle.net/10356/160069 10.1016/j.imavis.2020.104086 2-s2.0-85098474390 106 104086 en MOE2016-T2-1-042(S) Image and Vision Computing © 2020 Elsevier B.V. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering
Ordinal Classification
Salient Object Grading
spellingShingle Engineering::Computer science and engineering
Ordinal Classification
Salient Object Grading
Liu, Yanzhu
Wang, Yanan
Kong, Adams Wai Kin
Pixel-wise ordinal classification for salient object grading
description Driven by business intelligence applications for rating attraction of products in shops, a new problem — salient object grading is studied in this paper. In computer vision, plenty of salient object detection approaches have been proposed, while most existing studies detect objects in a binary manner: salient or not. This paper focuses on a new problem setting that requires detecting all salient objects and categorizing them into different salient levels. Based on that, a pixel-wise ordinal classification method is proposed. It consists of a multi-resolution saliency detector which detects and segments objects, an ordinal classifier which grades pixels into different salient levels, and a binary saliency enhancer which sharpens the difference between non-saliency and all other salient levels. Two new image datasets with salient level labels are constructed. Experimental results demonstrate that, on the one hand, the proposed method provides effective salient level predictions and on the other hand, offers very comparable performance with state-of-the-art salient object detection methods in the traditional problem setting.
author2 School of Computer Science and Engineering
author_facet School of Computer Science and Engineering
Liu, Yanzhu
Wang, Yanan
Kong, Adams Wai Kin
format Article
author Liu, Yanzhu
Wang, Yanan
Kong, Adams Wai Kin
author_sort Liu, Yanzhu
title Pixel-wise ordinal classification for salient object grading
title_short Pixel-wise ordinal classification for salient object grading
title_full Pixel-wise ordinal classification for salient object grading
title_fullStr Pixel-wise ordinal classification for salient object grading
title_full_unstemmed Pixel-wise ordinal classification for salient object grading
title_sort pixel-wise ordinal classification for salient object grading
publishDate 2022
url https://hdl.handle.net/10356/160069
_version_ 1738844790413328384