Global Guided Cross-Modal Cross-Scale Network for RGB-D Salient Object Detection
RGB-D saliency detection aims to accurately localize salient regions using the complementary information of a depth map. Global contexts carried by the deep layer are key to salient objection detection, but they are diluted when transferred to shallower layers. Besides, depth maps may contain mislea...
| Principais autores: | , , |
|---|---|
| Formato: | Artigo |
| Idioma: | English |
| Publicado em: |
MDPI AG
2023-08-01
|
| coleção: | Sensors |
| Assuntos: | |
| Acesso em linha: | https://www.mdpi.com/1424-8220/23/16/7221 |
| _version_ | 1827728695833395200 |
|---|---|
| author | Shuaihui Wang Fengyi Jiang Boqian Xu |
| author_facet | Shuaihui Wang Fengyi Jiang Boqian Xu |
| author_sort | Shuaihui Wang |
| collection | DOAJ |
| description | RGB-D saliency detection aims to accurately localize salient regions using the complementary information of a depth map. Global contexts carried by the deep layer are key to salient objection detection, but they are diluted when transferred to shallower layers. Besides, depth maps may contain misleading information due to the depth sensors. To tackle these issues, in this paper, we propose a new cross-modal cross-scale network for RGB-D salient object detection, where the global context information provides global guidance to boost performance in complex scenarios. First, we introduce a global guided cross-modal and cross-scale module named G<sup>2</sup>CMCSM to realize global guided cross-modal cross-scale fusion. Then, we employ feature refinement modules for progressive refinement in a coarse-to-fine manner. In addition, we adopt a hybrid loss function to supervise the training of G<sup>2</sup>CMCSNet over different scales. With all these modules working together, G<sup>2</sup>CMCSNet effectively enhances both salient object details and salient object localization. Extensive experiments on challenging benchmark datasets demonstrate that our G<sup>2</sup>CMCSNet outperforms existing state-of-the-art methods. |
| first_indexed | 2024-03-10T23:35:38Z |
| format | Article |
| id | doaj.art-6b8a3b709164402aa1e68cbf099bab72 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T23:35:38Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-6b8a3b709164402aa1e68cbf099bab722023-11-19T02:58:31ZengMDPI AGSensors1424-82202023-08-012316722110.3390/s23167221Global Guided Cross-Modal Cross-Scale Network for RGB-D Salient Object DetectionShuaihui Wang0Fengyi Jiang1Boqian Xu2Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaRGB-D saliency detection aims to accurately localize salient regions using the complementary information of a depth map. Global contexts carried by the deep layer are key to salient objection detection, but they are diluted when transferred to shallower layers. Besides, depth maps may contain misleading information due to the depth sensors. To tackle these issues, in this paper, we propose a new cross-modal cross-scale network for RGB-D salient object detection, where the global context information provides global guidance to boost performance in complex scenarios. First, we introduce a global guided cross-modal and cross-scale module named G<sup>2</sup>CMCSM to realize global guided cross-modal cross-scale fusion. Then, we employ feature refinement modules for progressive refinement in a coarse-to-fine manner. In addition, we adopt a hybrid loss function to supervise the training of G<sup>2</sup>CMCSNet over different scales. With all these modules working together, G<sup>2</sup>CMCSNet effectively enhances both salient object details and salient object localization. Extensive experiments on challenging benchmark datasets demonstrate that our G<sup>2</sup>CMCSNet outperforms existing state-of-the-art methods.https://www.mdpi.com/1424-8220/23/16/7221RGB-D salient object detectionglobal guidancecross-modal cross-scale fusion |
| spellingShingle | Shuaihui Wang Fengyi Jiang Boqian Xu Global Guided Cross-Modal Cross-Scale Network for RGB-D Salient Object Detection Sensors RGB-D salient object detection global guidance cross-modal cross-scale fusion |
| title | Global Guided Cross-Modal Cross-Scale Network for RGB-D Salient Object Detection |
| title_full | Global Guided Cross-Modal Cross-Scale Network for RGB-D Salient Object Detection |
| title_fullStr | Global Guided Cross-Modal Cross-Scale Network for RGB-D Salient Object Detection |
| title_full_unstemmed | Global Guided Cross-Modal Cross-Scale Network for RGB-D Salient Object Detection |
| title_short | Global Guided Cross-Modal Cross-Scale Network for RGB-D Salient Object Detection |
| title_sort | global guided cross modal cross scale network for rgb d salient object detection |
| topic | RGB-D salient object detection global guidance cross-modal cross-scale fusion |
| url | https://www.mdpi.com/1424-8220/23/16/7221 |
| work_keys_str_mv | AT shuaihuiwang globalguidedcrossmodalcrossscalenetworkforrgbdsalientobjectdetection AT fengyijiang globalguidedcrossmodalcrossscalenetworkforrgbdsalientobjectdetection AT boqianxu globalguidedcrossmodalcrossscalenetworkforrgbdsalientobjectdetection |