FPIseg: Iterative segmentation network based on feature pyramid for few‐shot segmentation
Abstract Few‐shot segmentation (FSS) enables rapid adaptation to the segmentation task of unseen‐classes object based on a few labelled support samples. Currently, the focal point of research in the FSS field is to align features between support and query images, aiming to improve the segmentation p...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-11-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.12898 |
| _version_ | 1797637254211960832 |
|---|---|
| author | Ronggui Wang Cong Yang Juan Yang Lixia Xue |
| author_facet | Ronggui Wang Cong Yang Juan Yang Lixia Xue |
| author_sort | Ronggui Wang |
| collection | DOAJ |
| description | Abstract Few‐shot segmentation (FSS) enables rapid adaptation to the segmentation task of unseen‐classes object based on a few labelled support samples. Currently, the focal point of research in the FSS field is to align features between support and query images, aiming to improve the segmentation performance. However, most existing FSS methods implement such support/query alignment by solely leveraging middle‐level feature for generalization, ignoring the category semantic information contained in high‐level feature, while pooling operation inevitably lose spatial information of the feature. To alleviate these issues, the authors propose the Iterative Segmentation Network Based on Feature Pyramid (FPIseg), which mainly consists of three modules: Feature Pyramid Fusion Module (FPFM), Region Feature Enhancement Module (RFEM), and Iterative Optimization Segmentation Module (IOSM). Firstly, FPFM fully utilizes the foreground information from the support image to implement support/query alignment under multi‐scale, multi‐level semantic backgrounds. Secondly, RFEM enhances the foreground detail information of aligned feature to improve generalization ability. Finally, ISOM iteratively segments the query image to optimize the prediction result and improve segmentation performance. Extensive experiments on the PASCAL‐5i and COCO‐20i datasets show that FPIseg achieves considerable segmentation performance under both 1‐shot and 5‐shot settings. |
| first_indexed | 2024-03-11T12:46:40Z |
| format | Article |
| id | doaj.art-6766f20c703b4c36bfab5dda084f223d |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-03-11T12:46:40Z |
| publishDate | 2023-11-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-6766f20c703b4c36bfab5dda084f223d2023-11-05T03:33:28ZengWileyIET Image Processing1751-96591751-96672023-11-0117133801381410.1049/ipr2.12898FPIseg: Iterative segmentation network based on feature pyramid for few‐shot segmentationRonggui Wang0Cong Yang1Juan Yang2Lixia Xue3School of Computer and Information Hefei University of Technology HefeiChinaSchool of Computer and Information Hefei University of Technology HefeiChinaSchool of Computer and Information Hefei University of Technology HefeiChinaSchool of Computer and Information Hefei University of Technology HefeiChinaAbstract Few‐shot segmentation (FSS) enables rapid adaptation to the segmentation task of unseen‐classes object based on a few labelled support samples. Currently, the focal point of research in the FSS field is to align features between support and query images, aiming to improve the segmentation performance. However, most existing FSS methods implement such support/query alignment by solely leveraging middle‐level feature for generalization, ignoring the category semantic information contained in high‐level feature, while pooling operation inevitably lose spatial information of the feature. To alleviate these issues, the authors propose the Iterative Segmentation Network Based on Feature Pyramid (FPIseg), which mainly consists of three modules: Feature Pyramid Fusion Module (FPFM), Region Feature Enhancement Module (RFEM), and Iterative Optimization Segmentation Module (IOSM). Firstly, FPFM fully utilizes the foreground information from the support image to implement support/query alignment under multi‐scale, multi‐level semantic backgrounds. Secondly, RFEM enhances the foreground detail information of aligned feature to improve generalization ability. Finally, ISOM iteratively segments the query image to optimize the prediction result and improve segmentation performance. Extensive experiments on the PASCAL‐5i and COCO‐20i datasets show that FPIseg achieves considerable segmentation performance under both 1‐shot and 5‐shot settings.https://doi.org/10.1049/ipr2.12898attention mechanismfeature engineeringfeature pyramid networkfew‐shot semantic segmentationprototype network |
| spellingShingle | Ronggui Wang Cong Yang Juan Yang Lixia Xue FPIseg: Iterative segmentation network based on feature pyramid for few‐shot segmentation IET Image Processing attention mechanism feature engineering feature pyramid network few‐shot semantic segmentation prototype network |
| title | FPIseg: Iterative segmentation network based on feature pyramid for few‐shot segmentation |
| title_full | FPIseg: Iterative segmentation network based on feature pyramid for few‐shot segmentation |
| title_fullStr | FPIseg: Iterative segmentation network based on feature pyramid for few‐shot segmentation |
| title_full_unstemmed | FPIseg: Iterative segmentation network based on feature pyramid for few‐shot segmentation |
| title_short | FPIseg: Iterative segmentation network based on feature pyramid for few‐shot segmentation |
| title_sort | fpiseg iterative segmentation network based on feature pyramid for few shot segmentation |
| topic | attention mechanism feature engineering feature pyramid network few‐shot semantic segmentation prototype network |
| url | https://doi.org/10.1049/ipr2.12898 |
| work_keys_str_mv | AT rongguiwang fpisegiterativesegmentationnetworkbasedonfeaturepyramidforfewshotsegmentation AT congyang fpisegiterativesegmentationnetworkbasedonfeaturepyramidforfewshotsegmentation AT juanyang fpisegiterativesegmentationnetworkbasedonfeaturepyramidforfewshotsegmentation AT lixiaxue fpisegiterativesegmentationnetworkbasedonfeaturepyramidforfewshotsegmentation |