Hilbert Curve-Encoded Rotation-Equivariant Oriented Object Detector with Locality-Preserving Spatial Mapping

Arbitrary-Oriented Object Detection (AOOD) has found broad applications in embodied intelligence, autonomous driving, and satellite remote sensing. However, current AOOD frameworks face challenges in ineffective feature extraction and orientation regression inaccuracy. Inspired by Hilbert curve's intrinsic locality-preserving property, we propose a flexible Hilbert curve-Encoded Rotation-Equivariant Oriented Object Detector (HERO-Det). Our innovations include: (i) a novel Hilbert curve traversal convolution paradigm with a dimensionality reduction scheme, which employs locality-preserving spatial filling curves for feature transformation, (ii) a Hilbert pyramid transformer enabling hierarchical construction of multi-scale feature sequences through space-folding operations, as well as (iii) an orientation-adaptive prediction head that decouples rotation-equivariant regression features from invariant classification cues to resolve orientation regression dilemmas in two-stage detectors. Extensive experiments show HERO-Det achieves state-of-the-art performance on AOOD benchmarks, with mAP of 79.56%, 90.64%, 90.10%, and 80.47% on DOTA, HRSC2016, SSDD, and HRSID, respectively. Performance gains in cross-task validation further demonstrate the versatility of our method to diverse vision tasks, such as medical image segmentation and 3D object detection.

Authors

• Qi Ming
• Liuqian Wang School of Cyber Science and Engineering, Zhengzhou University
• Juan Fang College of Computer Science, Beijing University of Technology
• Xudong Zhao
• Yucheng Xu Hong Kong University of Science and Technology (Guangzhou)
• Ziyi Teng College of Computer Science, Beijing University of Technology
• Yue Zhou Shanghai University
• Xiaoxi Hu State Key Laboratory of Intelligent Green Vehicle and Mobility, Tsinghua University
• Xiaohan Zhang South China University of Technology
• Yufei Guo Intelligent Science & Technology Academy of CASIC

Keywords

No keywords are indexed for this paper.