FSOD4RSI: Few-Shot Object Detection for Remote Sensing Images via Features Aggregation and Scale Attention
Due to the continuous development of few-shot learning, there have been notable advancements in methods for few-shot object detection in recent years. However, most existing methods in this domain primarily focus on natural images, neglecting the challenges posed by variations in object scales, whic...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10423123/ |
| _version_ | 1797299848524857344 |
|---|---|
| author | Honghao Gao Shuping Wu Ye Wang Jung Yoon Kim Yueshen Xu |
| author_facet | Honghao Gao Shuping Wu Ye Wang Jung Yoon Kim Yueshen Xu |
| author_sort | Honghao Gao |
| collection | DOAJ |
| description | Due to the continuous development of few-shot learning, there have been notable advancements in methods for few-shot object detection in recent years. However, most existing methods in this domain primarily focus on natural images, neglecting the challenges posed by variations in object scales, which are usually encountered in remote sensing images. This article proposes a new few-shot object detection model designed to handle the issue of object scale variation in remote sensing images. Our developed model has two essential parts: a feature aggregation module (FAM) and a scale-aware attention module (SAM). Considering the few-shot features of remote sensing images, we designed the FAM to improve the support and query features through channel multiplication operations utilizing a feature pyramid network and a transformer encoder. The created FAM better extracts the global features of remote sensing images and enhances the significant feature representation of few-shot remote sensing objects. In addition, we design the SAM to address the scale variation problems that frequently occur in remote sensing images. By employing multiscale convolutions, the SAM enables the acquisition of contextual features while adapting to objects of varying scales. Extensive experiments were conducted on benchmark datasets, including NWPU VHR-10 and DIOR datasets, and the results show that our model indeed addresses the challenges posed by object scale variation and improves the applicability of few-shot object detection in the remote sensing domain. |
| first_indexed | 2024-03-07T22:57:36Z |
| format | Article |
| id | doaj.art-cb7597c5ee904b5c9b67f86a6efd36c6 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-07T22:57:36Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-cb7597c5ee904b5c9b67f86a6efd36c62024-02-23T00:00:18ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-01174784479610.1109/JSTARS.2024.336274810423123FSOD4RSI: Few-Shot Object Detection for Remote Sensing Images via Features Aggregation and Scale AttentionHonghao Gao0https://orcid.org/0000-0001-6861-9684Shuping Wu1https://orcid.org/0009-0006-0454-2706Ye Wang2https://orcid.org/0000-0003-1454-2161Jung Yoon Kim3https://orcid.org/0000-0002-2396-9514Yueshen Xu4https://orcid.org/0000-0002-2396-9514School of Computer Engineering and Science, Shanghai University, Shanghai, ChinaSchool of Computer Engineering and Science, Shanghai University, Shanghai, ChinaSchool of Computer Engineering and Science, Shanghai University, Shanghai, ChinaCollege of Future Industry, Gachon University, Seongnam, 13120, South KoreaSchool of Computer Science and Technology, Xidian University, Xi'an, ChinaDue to the continuous development of few-shot learning, there have been notable advancements in methods for few-shot object detection in recent years. However, most existing methods in this domain primarily focus on natural images, neglecting the challenges posed by variations in object scales, which are usually encountered in remote sensing images. This article proposes a new few-shot object detection model designed to handle the issue of object scale variation in remote sensing images. Our developed model has two essential parts: a feature aggregation module (FAM) and a scale-aware attention module (SAM). Considering the few-shot features of remote sensing images, we designed the FAM to improve the support and query features through channel multiplication operations utilizing a feature pyramid network and a transformer encoder. The created FAM better extracts the global features of remote sensing images and enhances the significant feature representation of few-shot remote sensing objects. In addition, we design the SAM to address the scale variation problems that frequently occur in remote sensing images. By employing multiscale convolutions, the SAM enables the acquisition of contextual features while adapting to objects of varying scales. Extensive experiments were conducted on benchmark datasets, including NWPU VHR-10 and DIOR datasets, and the results show that our model indeed addresses the challenges posed by object scale variation and improves the applicability of few-shot object detection in the remote sensing domain.https://ieeexplore.ieee.org/document/10423123/Attention mechanismfeature aggregationfew-shot learningobject detectionremote sensing images |
| spellingShingle | Honghao Gao Shuping Wu Ye Wang Jung Yoon Kim Yueshen Xu FSOD4RSI: Few-Shot Object Detection for Remote Sensing Images via Features Aggregation and Scale Attention IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Attention mechanism feature aggregation few-shot learning object detection remote sensing images |
| title | FSOD4RSI: Few-Shot Object Detection for Remote Sensing Images via Features Aggregation and Scale Attention |
| title_full | FSOD4RSI: Few-Shot Object Detection for Remote Sensing Images via Features Aggregation and Scale Attention |
| title_fullStr | FSOD4RSI: Few-Shot Object Detection for Remote Sensing Images via Features Aggregation and Scale Attention |
| title_full_unstemmed | FSOD4RSI: Few-Shot Object Detection for Remote Sensing Images via Features Aggregation and Scale Attention |
| title_short | FSOD4RSI: Few-Shot Object Detection for Remote Sensing Images via Features Aggregation and Scale Attention |
| title_sort | fsod4rsi few shot object detection for remote sensing images via features aggregation and scale attention |
| topic | Attention mechanism feature aggregation few-shot learning object detection remote sensing images |
| url | https://ieeexplore.ieee.org/document/10423123/ |
| work_keys_str_mv | AT honghaogao fsod4rsifewshotobjectdetectionforremotesensingimagesviafeaturesaggregationandscaleattention AT shupingwu fsod4rsifewshotobjectdetectionforremotesensingimagesviafeaturesaggregationandscaleattention AT yewang fsod4rsifewshotobjectdetectionforremotesensingimagesviafeaturesaggregationandscaleattention AT jungyoonkim fsod4rsifewshotobjectdetectionforremotesensingimagesviafeaturesaggregationandscaleattention AT yueshenxu fsod4rsifewshotobjectdetectionforremotesensingimagesviafeaturesaggregationandscaleattention |