References & Citations
Computer Science > Computer Vision and Pattern Recognition
Title: OSKDet: Towards Orientation-sensitive Keypoint Localization for Rotated Object Detection
(Submitted on 18 Apr 2021 (v1), last revised 28 Jun 2021 (this version, v2))
Abstract: Rotated object detection is a challenging issue of computer vision field. Loss of spatial information and confusion of parametric order have been the bottleneck for rotated detection accuracy. In this paper, we propose an orientation-sensitive keypoint based rotated detector OSKDet. We adopt a set of keypoints to characterize the target and predict the keypoint heatmap on ROI to form a rotated target. By proposing the orientation-sensitive heatmap, OSKDet could learn the shape and direction of rotated target implicitly and has stronger modeling capabilities for target representation, which improves the localization accuracy and acquires high quality detection results. To extract highly effective features at border areas, we design a rotation-aware deformable convolution module. Furthermore, we explore a new keypoint reorder algorithm and feature fusion module based on the angle distribution to eliminate the confusion of keypoint order. Experimental results on several public benchmarks show the state-of-the-art performance of OSKDet. Specifically, we achieve an AP of 77.81% on DOTA, 89.91% on HRSC2016, and 97.18% on UCAS-AOD, respectively.
Submission history
From: Dongchen Lu [view email][v1] Sun, 18 Apr 2021 03:40:52 GMT (20540kb,D)
[v2] Mon, 28 Jun 2021 14:57:11 GMT (10266kb,D)
Link back to: arXiv, form interface, contact.