Abstract: Domain adaptive object detection methods for remote sensing images (RSIs) typically rely on large-scale target-domain data and multiepoch offline adaption training. However, the wide ...

Abstract: Traditional object detection models often lose the detailed outline information of the object. To address this problem, we propose the Fourier Series Object Detection (FSD). It encodes the ...

This repository provides an end-to-end pipeline to detect and help prevent deepfake images and videos. It includes training code, inference scripts, visualization (Grad-CAM heatmaps), and a Streamlit ...

propose an instance-level alignment module (PDFA) to dynamically adjust the focus of model based on the prediction-discrepancy feedback; develop an image-level alignment module (UFOA) to adaptively ...