Few-Shot Object Detection on Remote Sensing Images via Shared Attention Module and Balanced Fine-Tuning Strategy
Few-shot object detection is a recently emerging branch in the field of computer vision. Recent research studies have proposed several effective methods for object detection with few samples. However, their performances are limited when applied to remote sensing images. In this article, we specifica...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/19/3816 |
| _version_ | 1797515870848679936 |
|---|---|
| author | Xu Huang Bokun He Ming Tong Dingwen Wang Chu He |
| author_facet | Xu Huang Bokun He Ming Tong Dingwen Wang Chu He |
| author_sort | Xu Huang |
| collection | DOAJ |
| description | Few-shot object detection is a recently emerging branch in the field of computer vision. Recent research studies have proposed several effective methods for object detection with few samples. However, their performances are limited when applied to remote sensing images. In this article, we specifically analyze the characteristics of remote sensing images and propose a few-shot fine-tuning network with a shared attention module (SAM) to adapt to detecting remote sensing objects, which have large size variations. In our SAM, multi-attention maps are computed in the base training stage and shared with the feature extractor in the few-shot fine-tuning stage as prior knowledge to help better locate novel class objects with few samples. Moreover, we design a new few-shot fine-tuning stage with a balanced fine-tuning strategy (BFS), which helps in mitigating the severe imbalance between the number of novel class samples and base class samples caused by the few-shot settings to improve the classification accuracy. We have conducted experiments on two remote sensing datasets (NWPU VHR-10 and DIOR), and the excellent results demonstrate that our method makes full use of the advantages of few-shot learning and the characteristics of remote sensing images to enhance the few-shot detection performance. |
| first_indexed | 2024-03-10T06:53:21Z |
| format | Article |
| id | doaj.art-bbc8b61271984381b7b6808b08112e84 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T06:53:21Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-bbc8b61271984381b7b6808b08112e842023-11-22T16:41:16ZengMDPI AGRemote Sensing2072-42922021-09-011319381610.3390/rs13193816Few-Shot Object Detection on Remote Sensing Images via Shared Attention Module and Balanced Fine-Tuning StrategyXu Huang0Bokun He1Ming Tong2Dingwen Wang3Chu He4School of Computer Science, Wuhan University, Wuhan 430072, ChinaSchool of Electronic Information, Wuhan University, Wuhan 430072, ChinaSchool of Electronic Information, Wuhan University, Wuhan 430072, ChinaSchool of Computer Science, Wuhan University, Wuhan 430072, ChinaSchool of Electronic Information, Wuhan University, Wuhan 430072, ChinaFew-shot object detection is a recently emerging branch in the field of computer vision. Recent research studies have proposed several effective methods for object detection with few samples. However, their performances are limited when applied to remote sensing images. In this article, we specifically analyze the characteristics of remote sensing images and propose a few-shot fine-tuning network with a shared attention module (SAM) to adapt to detecting remote sensing objects, which have large size variations. In our SAM, multi-attention maps are computed in the base training stage and shared with the feature extractor in the few-shot fine-tuning stage as prior knowledge to help better locate novel class objects with few samples. Moreover, we design a new few-shot fine-tuning stage with a balanced fine-tuning strategy (BFS), which helps in mitigating the severe imbalance between the number of novel class samples and base class samples caused by the few-shot settings to improve the classification accuracy. We have conducted experiments on two remote sensing datasets (NWPU VHR-10 and DIOR), and the excellent results demonstrate that our method makes full use of the advantages of few-shot learning and the characteristics of remote sensing images to enhance the few-shot detection performance.https://www.mdpi.com/2072-4292/13/19/3816object detectionfew-shot learningremote sensing imagesattention mechanism |
| spellingShingle | Xu Huang Bokun He Ming Tong Dingwen Wang Chu He Few-Shot Object Detection on Remote Sensing Images via Shared Attention Module and Balanced Fine-Tuning Strategy Remote Sensing object detection few-shot learning remote sensing images attention mechanism |
| title | Few-Shot Object Detection on Remote Sensing Images via Shared Attention Module and Balanced Fine-Tuning Strategy |
| title_full | Few-Shot Object Detection on Remote Sensing Images via Shared Attention Module and Balanced Fine-Tuning Strategy |
| title_fullStr | Few-Shot Object Detection on Remote Sensing Images via Shared Attention Module and Balanced Fine-Tuning Strategy |
| title_full_unstemmed | Few-Shot Object Detection on Remote Sensing Images via Shared Attention Module and Balanced Fine-Tuning Strategy |
| title_short | Few-Shot Object Detection on Remote Sensing Images via Shared Attention Module and Balanced Fine-Tuning Strategy |
| title_sort | few shot object detection on remote sensing images via shared attention module and balanced fine tuning strategy |
| topic | object detection few-shot learning remote sensing images attention mechanism |
| url | https://www.mdpi.com/2072-4292/13/19/3816 |
| work_keys_str_mv | AT xuhuang fewshotobjectdetectiononremotesensingimagesviasharedattentionmoduleandbalancedfinetuningstrategy AT bokunhe fewshotobjectdetectiononremotesensingimagesviasharedattentionmoduleandbalancedfinetuningstrategy AT mingtong fewshotobjectdetectiononremotesensingimagesviasharedattentionmoduleandbalancedfinetuningstrategy AT dingwenwang fewshotobjectdetectiononremotesensingimagesviasharedattentionmoduleandbalancedfinetuningstrategy AT chuhe fewshotobjectdetectiononremotesensingimagesviasharedattentionmoduleandbalancedfinetuningstrategy |