Multiscale Deep Network with Centerness-Aware Loss for Salient Object Detection
Deep encoder-decoder networks have been adopted for saliency detection and achieved state-of-the-art performance. However, most existing saliency models usually fail to detect very small salient objects. In this paper, we propose a multitask architecture, M2Net, and a novel centerness-aware loss for...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-01-01
|
| Series: | Advances in Multimedia |
| Online Access: | http://dx.doi.org/10.1155/2022/2243927 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832563722103554048 |
|---|---|
| author | Liangliang Duan |
| author_facet | Liangliang Duan |
| author_sort | Liangliang Duan |
| collection | DOAJ |
| description | Deep encoder-decoder networks have been adopted for saliency detection and achieved state-of-the-art performance. However, most existing saliency models usually fail to detect very small salient objects. In this paper, we propose a multitask architecture, M2Net, and a novel centerness-aware loss for salient object detection. The proposed M2Net aims to solve saliency prediction and centerness prediction simultaneously. Specifically, the network architecture is composed of a bottom-up encoder module, top-down decoder module, and centerness prediction module. In addition, different from binary cross entropy, the proposed centerness-aware loss can guide the proposed M2Net to uniformly highlight the entire salient regions with well-defined object boundaries. Experimental results on five benchmark saliency datasets demonstrate that M2Net outperforms state-of-the-art methods on different evaluation metrics. |
| format | Article |
| id | doaj-art-a11326ac94a542f8ac00137714ff9049 |
| institution | Kabale University |
| issn | 1687-5699 |
| language | English |
| publishDate | 2022-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Advances in Multimedia |
| spelling | doaj-art-a11326ac94a542f8ac00137714ff90492025-02-03T01:12:53ZengWileyAdvances in Multimedia1687-56992022-01-01202210.1155/2022/2243927Multiscale Deep Network with Centerness-Aware Loss for Salient Object DetectionLiangliang Duan0Qingdao University of TechnologyDeep encoder-decoder networks have been adopted for saliency detection and achieved state-of-the-art performance. However, most existing saliency models usually fail to detect very small salient objects. In this paper, we propose a multitask architecture, M2Net, and a novel centerness-aware loss for salient object detection. The proposed M2Net aims to solve saliency prediction and centerness prediction simultaneously. Specifically, the network architecture is composed of a bottom-up encoder module, top-down decoder module, and centerness prediction module. In addition, different from binary cross entropy, the proposed centerness-aware loss can guide the proposed M2Net to uniformly highlight the entire salient regions with well-defined object boundaries. Experimental results on five benchmark saliency datasets demonstrate that M2Net outperforms state-of-the-art methods on different evaluation metrics.http://dx.doi.org/10.1155/2022/2243927 |
| spellingShingle | Liangliang Duan Multiscale Deep Network with Centerness-Aware Loss for Salient Object Detection Advances in Multimedia |
| title | Multiscale Deep Network with Centerness-Aware Loss for Salient Object Detection |
| title_full | Multiscale Deep Network with Centerness-Aware Loss for Salient Object Detection |
| title_fullStr | Multiscale Deep Network with Centerness-Aware Loss for Salient Object Detection |
| title_full_unstemmed | Multiscale Deep Network with Centerness-Aware Loss for Salient Object Detection |
| title_short | Multiscale Deep Network with Centerness-Aware Loss for Salient Object Detection |
| title_sort | multiscale deep network with centerness aware loss for salient object detection |
| url | http://dx.doi.org/10.1155/2022/2243927 |
| work_keys_str_mv | AT liangliangduan multiscaledeepnetworkwithcenternessawarelossforsalientobjectdetection |