Cross refinement network with edge detection for salient object detection
Abstract Salient object detection aims to identify the most attractive objects from images. However, their boundaries are typically of poor quality when predicted using available methods. One or multiple objects may also be left undetected if the image contains multiple objects. To solve these probl...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-09-01
|
| Series: | IET Signal Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/sil2.12041 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832546776223055872 |
|---|---|
| author | Junjiang Xiang Xiao Hu Jiayu Ding Xiangyue Tan Jiaxin Yang |
| author_facet | Junjiang Xiang Xiao Hu Jiayu Ding Xiangyue Tan Jiaxin Yang |
| author_sort | Junjiang Xiang |
| collection | DOAJ |
| description | Abstract Salient object detection aims to identify the most attractive objects from images. However, their boundaries are typically of poor quality when predicted using available methods. One or multiple objects may also be left undetected if the image contains multiple objects. To solve these problems, this article proposes the novel cross refinement network, which consists of a Res2Net‐based backbone network; a fusion network equipped with four convolutional block attention modules and four edge‐salient cross units; and a detection network with an edge enhancement unit and a residual refinement network (RNN). For RNN training, the rough salient maps generated using the DUTS‐TR dataset are treated as a special training dataset. Compared to existing methods, the proposed network can effectively detect all objects as well as improve the boundaries of the detected objects by performing experiments on five benchmark datasets. Based on the experimental results, the proposed network outperforms existing methods both objectively and subjectively. |
| format | Article |
| id | doaj-art-f1283173cd1246a29ed2c314af17f355 |
| institution | Kabale University |
| issn | 1751-9675 1751-9683 |
| language | English |
| publishDate | 2021-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Signal Processing |
| spelling | doaj-art-f1283173cd1246a29ed2c314af17f3552025-02-03T06:47:26ZengWileyIET Signal Processing1751-96751751-96832021-09-0115742543610.1049/sil2.12041Cross refinement network with edge detection for salient object detectionJunjiang Xiang0Xiao Hu1Jiayu Ding2Xiangyue Tan3Jiaxin Yang4School of Mechanical and Electrical Engineering Guangzhou University Guangzhou ChinaSchool of Mechanical and Electrical Engineering Guangzhou University Guangzhou ChinaChina Huangpu Research & Graduate School of Guangzhou University Guangzhou ChinaChina Huangpu Research & Graduate School of Guangzhou University Guangzhou ChinaChina Huangpu Research & Graduate School of Guangzhou University Guangzhou ChinaAbstract Salient object detection aims to identify the most attractive objects from images. However, their boundaries are typically of poor quality when predicted using available methods. One or multiple objects may also be left undetected if the image contains multiple objects. To solve these problems, this article proposes the novel cross refinement network, which consists of a Res2Net‐based backbone network; a fusion network equipped with four convolutional block attention modules and four edge‐salient cross units; and a detection network with an edge enhancement unit and a residual refinement network (RNN). For RNN training, the rough salient maps generated using the DUTS‐TR dataset are treated as a special training dataset. Compared to existing methods, the proposed network can effectively detect all objects as well as improve the boundaries of the detected objects by performing experiments on five benchmark datasets. Based on the experimental results, the proposed network outperforms existing methods both objectively and subjectively.https://doi.org/10.1049/sil2.12041edge detectionlearning (artificial intelligence)object detectionrecurrent neural nets |
| spellingShingle | Junjiang Xiang Xiao Hu Jiayu Ding Xiangyue Tan Jiaxin Yang Cross refinement network with edge detection for salient object detection IET Signal Processing edge detection learning (artificial intelligence) object detection recurrent neural nets |
| title | Cross refinement network with edge detection for salient object detection |
| title_full | Cross refinement network with edge detection for salient object detection |
| title_fullStr | Cross refinement network with edge detection for salient object detection |
| title_full_unstemmed | Cross refinement network with edge detection for salient object detection |
| title_short | Cross refinement network with edge detection for salient object detection |
| title_sort | cross refinement network with edge detection for salient object detection |
| topic | edge detection learning (artificial intelligence) object detection recurrent neural nets |
| url | https://doi.org/10.1049/sil2.12041 |
| work_keys_str_mv | AT junjiangxiang crossrefinementnetworkwithedgedetectionforsalientobjectdetection AT xiaohu crossrefinementnetworkwithedgedetectionforsalientobjectdetection AT jiayuding crossrefinementnetworkwithedgedetectionforsalientobjectdetection AT xiangyuetan crossrefinementnetworkwithedgedetectionforsalientobjectdetection AT jiaxinyang crossrefinementnetworkwithedgedetectionforsalientobjectdetection |