MidNet: An Anchor-and-Angle-Free Detector for Oriented Ship Detection in Aerial Images
This work addresses the problem of accurate ship detection for remote sensing applications, offering a novel approach that reduces sensitivity to angular regression and hyper-parameter tuning, though it is incremental in improving existing detection paradigms.
The paper tackles the challenge of oriented ship detection in aerial images by introducing MidNet, an anchor-and-angle-free detector that uses center and midpoints for object encoding, achieving APs of 90.52% on HRSC2016 and 86.50% on FGSD2021, outperforming state-of-the-art methods.
Ship detection in aerial images remains an active yet challenging task due to arbitrary object orientation and complex background from a bird's-eye perspective. Most of the existing methods rely on angular prediction or predefined anchor boxes, making these methods highly sensitive to unstable angular regression and excessive hyper-parameter setting. To address these issues, we replace the angular-based object encoding with an anchor-and-angle-free paradigm, and propose a novel detector deploying a center and four midpoints for encoding each oriented object, namely MidNet. MidNet designs a symmetrical deformable convolution customized for enhancing the midpoints of ships, then the center and midpoints for an identical ship are adaptively matched by predicting corresponding centripetal shift and matching radius. Finally, a concise analytical geometry algorithm is proposed to refine the centers and midpoints step-wisely for building precise oriented bounding boxes. On two public ship detection datasets, HRSC2016 and FGSD2021, MidNet outperforms the state-of-the-art detectors by achieving APs of 90.52% and 86.50%. Additionally, MidNet obtains competitive results in the ship detection of DOTA.