Abstract: A key issue of existing deep-learning-based object detection methods in remote sensing images is that they often struggle to differentiate the background and small object regions due to ...

Abstract: Object detection in aerial images has always been a challenging task due to the generally small size of the objects. Most current detectors prioritize the development of new detection ...

With the rapid development of marine resource exploitation and the increasing demand for underwater robot inspection, achieving reliable target perception in turbid, low-illumination, and spectrally ...

I tried using DINOv3 as the pre-trained model for the detector and encountered an issue. When defining the Transformer, self.reference_points（not two-stage） is initialized as follows: if two_stage: ...

Traffic monitoring plays a vital role in smart city infrastructure, road safety, and urban planning. Traditional detection systems, including earlier deep learning models, often struggle with ...