Hybrid Attention Asynchronous Cascade Network for Salient Object Detection
The highlighted area or object is defined as the salient region or salient object. For salient object detection, the main challenges are still the clarity of the boundary information of the salient object and the positioning accuracy of the salient object in the complex background, such as noise and...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/11/6/1389 |
| _version_ | 1827748866799173632 |
|---|---|
| author | Haiyan Yang Yongxin Chen Rui Chen Shuning Liu |
| author_facet | Haiyan Yang Yongxin Chen Rui Chen Shuning Liu |
| author_sort | Haiyan Yang |
| collection | DOAJ |
| description | The highlighted area or object is defined as the salient region or salient object. For salient object detection, the main challenges are still the clarity of the boundary information of the salient object and the positioning accuracy of the salient object in the complex background, such as noise and occlusion. To remedy these issues, it is proposed that the asynchronous cascade saliency detection algorithm based on a deep network, which is embedded in an encoder–decoder architecture. Moreover, the lightweight hybrid attention module is designed to obtain the explicit boundaries of salient regions. In order to effectively improve location information of salient objects, this paper adopts a bi-directional asynchronous cascade fusion strategy, which generates prediction maps with higher accuracy. The experimental results on five benchmark datasets show that the proposed network HACNet is on a par with the state of the art for image saliency datasets. |
| first_indexed | 2024-03-11T06:12:58Z |
| format | Article |
| id | doaj.art-026f1f575e0d4476bd5c7e8770e96b11 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-11T06:12:58Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-026f1f575e0d4476bd5c7e8770e96b112023-11-17T12:27:54ZengMDPI AGMathematics2227-73902023-03-01116138910.3390/math11061389Hybrid Attention Asynchronous Cascade Network for Salient Object DetectionHaiyan Yang0Yongxin Chen1Rui Chen2Shuning Liu3School of Information and Communication, Guilin University of Electronic Technology, Guilin 541004, ChinaSchool of Information and Communication, Guilin University of Electronic Technology, Guilin 541004, ChinaSchool of Information and Communication, Guilin University of Electronic Technology, Guilin 541004, ChinaSchool of Information and Communication, Guilin University of Electronic Technology, Guilin 541004, ChinaThe highlighted area or object is defined as the salient region or salient object. For salient object detection, the main challenges are still the clarity of the boundary information of the salient object and the positioning accuracy of the salient object in the complex background, such as noise and occlusion. To remedy these issues, it is proposed that the asynchronous cascade saliency detection algorithm based on a deep network, which is embedded in an encoder–decoder architecture. Moreover, the lightweight hybrid attention module is designed to obtain the explicit boundaries of salient regions. In order to effectively improve location information of salient objects, this paper adopts a bi-directional asynchronous cascade fusion strategy, which generates prediction maps with higher accuracy. The experimental results on five benchmark datasets show that the proposed network HACNet is on a par with the state of the art for image saliency datasets.https://www.mdpi.com/2227-7390/11/6/1389salient object detectionlightweight hybrid attention moduleasynchronous cascading strategyinteractive residual moduleparallel dilated convolution |
| spellingShingle | Haiyan Yang Yongxin Chen Rui Chen Shuning Liu Hybrid Attention Asynchronous Cascade Network for Salient Object Detection Mathematics salient object detection lightweight hybrid attention module asynchronous cascading strategy interactive residual module parallel dilated convolution |
| title | Hybrid Attention Asynchronous Cascade Network for Salient Object Detection |
| title_full | Hybrid Attention Asynchronous Cascade Network for Salient Object Detection |
| title_fullStr | Hybrid Attention Asynchronous Cascade Network for Salient Object Detection |
| title_full_unstemmed | Hybrid Attention Asynchronous Cascade Network for Salient Object Detection |
| title_short | Hybrid Attention Asynchronous Cascade Network for Salient Object Detection |
| title_sort | hybrid attention asynchronous cascade network for salient object detection |
| topic | salient object detection lightweight hybrid attention module asynchronous cascading strategy interactive residual module parallel dilated convolution |
| url | https://www.mdpi.com/2227-7390/11/6/1389 |
| work_keys_str_mv | AT haiyanyang hybridattentionasynchronouscascadenetworkforsalientobjectdetection AT yongxinchen hybridattentionasynchronouscascadenetworkforsalientobjectdetection AT ruichen hybridattentionasynchronouscascadenetworkforsalientobjectdetection AT shuningliu hybridattentionasynchronouscascadenetworkforsalientobjectdetection |