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 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.

[1]  Gang Wan,et al.  Object Detection in Optical Remote Sensing Images: A Survey and A New Benchmark , 2020, ISPRS Journal of Photogrammetry and Remote Sensing.

[2]  Jie Geng,et al.  Prototype Calibration with Feature Generation for Few-Shot Remote Sensing Image Scene Classification , 2021, Remote. Sens..

[3]  Junwei Han,et al.  Multi-class geospatial object detection and geographic image classification based on collection of part detectors , 2014 .

[4]  Zhenfeng Shao,et al.  Meta-FSEO: A Meta-Learning Fast Adaptation with Self-Supervised Embedding Optimization for Few-Shot Remote Sensing Scene Classification , 2021, Remote. Sens..

[5]  Dong Wang,et al.  Few-Shot Classification of Aerial Scene Images via Meta-Learning , 2021, Remote. Sens..

[6]  Kai Chen,et al.  MMDetection: Open MMLab Detection Toolbox and Benchmark , 2019, ArXiv.

[7]  Elif Sertel,et al.  Comparative Research on Deep Learning Approaches for Airplane Detection from Very High-Resolution Satellite Images , 2020, Remote. Sens..

[8]  Jing Zhang,et al.  Object Detection Based on Global-Local Saliency Constraint in Aerial Images , 2020, Remote. Sens..

[9]  Wei Guo,et al.  Geospatial Object Detection in High Resolution Satellite Images Based on Multi-Scale Convolutional Neural Network , 2018, Remote. Sens..

[10]  Oriol Vinyals,et al.  Matching Networks for One Shot Learning , 2016, NIPS.

[11]  Ali Farhadi,et al.  YOLOv3: An Incremental Improvement , 2018, ArXiv.

[12]  J. Niemeyer,et al.  Contextual classification of lidar data and building object detection in urban areas , 2014 .