A Mask Guided Oriented Object Detector Based on Rotated Size-Adaptive Tricube Kernel

Oriented object detection is an important research topic in remote sensing. The detection of oriented objects in remote sensing images remains a daunting challenge due to their complex backgrounds, various sizes, diverse aspect ratios, and especially arbitrary orientations. In recent years, keypoint-based anchor-free object detectors have demonstrated outstanding performance in this field. However, in current anchor-free detectors, object keypoints are primarily generated using the Gaussian kernel function, which assumes a circular form. This representation falls short in accurately conveying an object's size and orientation. To address the aforementioned issue, this article proposes a keypoint-based oriented object detector called MRSDet, which innovatively adopts the Tricube kernel, scales, and rotates it, to better generate the center keypoint heatmap of the object. Besides, to improve the model's detection performance on oriented objects and improve its ability to perceive object keypoints and boundary boxes, we also design a large receptive field mask (LRFM) module, which is based on large convolution kernel decomposition and semantic segmentation masks. Taking the box boundary-aware vectors (BBAVectors) method as a baseline, we conduct experiments on multiple types of remote sensing datasets such as HRSC2016, UCAS-AOD, and SSDD+ datasets to verify the effectiveness and generalizability of the proposed method.