Abstract:Unsupervised domain adaptation (UDA) is critical for remote sensing object detection in real applications, aiming to address the significant performance degradation issue caused by the domain gap between the source and target domain. This method achieves cross-domain alignment by leveraging the unlabeled target domain data, thus avoiding the expensive annotation cost. However, existing works mainly cope with convolutional neural network (CNN)-based object detectors, which are characterized by complex adversarial learning architecture and fail to accurately align the features in remote sensing images with sparsely allocated objects and inevitable background noise. Compared to CNN-based methods, the detection transformer (DETR) largely simplifies the object detection pipeline and demonstrates the great potential of its intrinsic characteristics of global relation modeling between any pixels. On this basis, we propose the first strong DETR-based baseline, remote sensing teacher, for UDA in remote sensing object detection. Specifically, the remote sensing teacher introduces an innovative learnable frequency-enhanced feature alignment (LFA) module. Within this module, we initially transform the features into frequency space to simplify the attention solver and effectively capture domain-specific information. Subsequently, the module significantly enhances the global feature representations of sparsely allocated objects by using a lightweight attention mechanism. Following this, the module incorporates learnable filters with a gated mechanism, enabling selective alignment of features in noisy backgrounds. In addition, the remote sensing teacher employs a self-adaptive pseudo-label assigner (SPA) that can automatically adjust the class-wise confidence threshold according to the model’s learning status, thereby enabling the generation of high-quality pseudo-labels in scenarios with a long-tailed distribution. Leveraging these pseudo-labels further mitigates the domain bias of the detector by establishing alignment at the label level. Extensive experimental results demonstrate the superior performance and generalization capabilities of our proposed remote sensing teacher in multiple remote sensing adaptation scenarios. The Code is released at https://github.com/h751410234/RemoteSensingTeacher.

CLT-Det: Correlation Learning Based on Transformer for Detecting Dense Objects in Remote Sensing Images

A Transformer-Based Object Detector with Coarse-Fine Crossing Representations

YOLO-DCTI: Small Object Detection in Remote Sensing Base on Contextual Transformer Enhancement

High-Resolution Network with Transformer Embedding Parallel Detection for Small Object Detection in Optical Remote Sensing Images

Remote Sensing Object Detection Based on Strong Feature Extraction and Prescreening Network

A Convolution with Transformer Attention Module Integrating Local and Global Features for Object Detection in Remote Sensing Based on YOLOv8n

MDCT: Multi-Kernel Dilated Convolution and Transformer for One-Stage Object Detection of Remote Sensing Images.

Transformer with Transfer CNN for Remote-Sensing-Image Object Detection

Remote Sensing Object Detection in Disturbed Environment Based on Improved RTMDet Network

C3TB-YOLOv5: integrated YOLOv5 with transformer for object detection in high-resolution remote sensing images

ADT-Det: Adaptive Dynamic Refined Single-Stage Transformer Detector for Arbitrary-Oriented Object Detection in Satellite Optical Imagery

Dynamic Sensing and Correlation Loss Detector for Small Object Detection in Remote Sensing Images

Structured Object-Level Relational Reasoning CNN-Based Target Detection Algorithm in a Remote Sensing Image

DETR-ORD: An Improved DETR Detector for Oriented Remote Sensing Object Detection with Feature Reconstruction and Dynamic Query

DFS-DETR: Detailed-Feature-Sensitive Detector for Small Object Detection in Aerial Images Using Transformer

CCDN-DETR: A Detection Transformer Based on Constrained Contrast Denoising for Multi-Class Synthetic Aperture Radar Object Detection

Transformer with large convolution kernel decoder network for salient object detection in optical remote sensing images

TransY-Net:Learning Fully Transformer Networks for Change Detection of Remote Sensing Images

Remote Sensing Teacher: Cross-Domain Detection Transformer with Learnable Frequency-Enhanced Feature Alignment in Remote Sensing Imagery

TransY-Net: Learning Fully Transformer Networks for Change Detection of Remote Sensing Images

QETR: A Query-Enhanced Transformer for Remote Sensing Image Object Detection