Filters








48,704 Hits in 6.9 sec

Anchor-free Oriented Proposal Generator for Object Detection [article]

Gong Cheng and Jiabao Wang and Ke Li and Xingxing Xie and Chunbo Lang and Yanqing Yao and Junwei Han
2022 arXiv   pre-print
In this paper, we propose a novel Anchor-free Oriented Proposal Generator (AOPG) that abandons horizontal box-related operations from the network architecture.  ...  AOPG first produces coarse oriented boxes by a Coarse Location Module (CLM) in an anchor-free manner and then refines them into high-quality oriented proposals.  ...  In this work, we propose a novel Anchor-free Oriented Proposal Generator (AOPG) by adopting the anchor-free scheme towards generating high-quality oriented proposals.  ... 
arXiv:2110.01931v2 fatcat:bxjbrlwynne4hnxkowfj6hsnju

Anchor Retouching via Model Interaction for Robust Object Detection in Aerial Images [article]

Dong Liang, Qixiang Geng, Zongqi Wei, Dmitry A. Vorontsov, Ekaterina L. Kim, Mingqiang Wei, Huiyu Zhou
2021 arXiv   pre-print
The proposed scheme supports both oriented and horizontal object detection tasks.  ...  Different from the other state-of-the-art techniques, the proposed network leverages a sample discriminator to realize interactive sample screening between an anchor-based unit and an anchor-free unit  ...  ACKNOWLEDGMENT The authors would like to thank Professor Sheng-Jun Huang from NUAA and Professor Gui-Song Xia from Wuhan University for their help discussion and comments.  ... 
arXiv:2112.06701v1 fatcat:qr6dfhr43vcifau7ywtasj3a4m

Axis Learning for Orientated Objects Detection in Aerial Images

Zhifeng Xiao, Linjun Qian, Weiping Shao, Xiaowei Tan, Kai Wang
2020 Remote Sensing  
In this article, we propose a new one-stage anchor-free method to detect orientated objects in per-pixel prediction fashion with less computational complexity.  ...  Most current methods for orientated object detection are anchor-based, which require considerable pre-defined anchors and are time consuming.  ...  In this paper, we propose a new one-stage anchor-free orientated objects detector for aerial images.  ... 
doi:10.3390/rs12060908 fatcat:ohce7ratdneqpgmvvjtzuozk6q

DAFNe: A One-Stage Anchor-Free Approach for Oriented Object Detection [article]

Steven Lang, Fabrizio Ventola, Kristian Kersting
2022 arXiv   pre-print
We present DAFNe, a Dense one-stage Anchor-Free deep Network for oriented object detection.  ...  With DAFNe we introduce an orientation-aware generalization of the center-ness function for arbitrarily oriented bounding boxes to down-weight low-quality predictions and a center-to-corner bounding box  ...  Acknowledgements This work was supported by the Federal Ministry of Education and Research (BMBF) Competence Center for AI and Labour "kompAKI" (FKZ 02L19C150), the German Science Foundation (DFG, German  ... 
arXiv:2109.06148v4 fatcat:lsiyerza6vf6hh3cs3onhiwjfi

Oriented R-CNN for Object Detection [article]

Xingxing Xie, Gong Cheng, Jiabao Wang, Xiwen Yao, Junwei Han
2021 arXiv   pre-print
To be specific, in the first stage, we propose an oriented Region Proposal Network (oriented RPN) that directly generates high-quality oriented proposals in a nearly cost-free manner.  ...  This work proposes an effective and simple oriented object detection framework, termed Oriented R-CNN, which is a general two-stage oriented detector with promising accuracy and efficiency.  ...  Pan et.al [28] devised a dynamic refinement network (DRN) for oriented object detection based on the anchor-free object detection method CenterNet [46] .  ... 
arXiv:2108.05699v1 fatcat:5tafjzxc5vfl5irzm7zscbob4u

Oriented Object Detection in Remote Sensing Images with Anchor-Free Oriented Region Proposal Network

Jianxiang Li, Yan Tian, Yiping Xu, Zili Zhang
2022 Remote Sensing  
We propose the Anchor-Free Oriented Region Proposal Network (AFO-RPN) to generate high-quality oriented proposals without enormous predefined anchors.  ...  To address the problem, this paper proposes a novel anchor-free two-stage oriented object detector.  ...  Most anchor-free oriented object detectors are inherent from CenterNet for high efficiency and generality, having achieved performance competitive with anchor-based detectors. For example, Pan et al.  ... 
doi:10.3390/rs14051246 fatcat:hsnpxrds45a5dbkgkfgbdrc43u

Oriented Objects as pairs of Middle Lines [article]

Haoran Wei, Yue Zhang, Zhonghan Chang, Hao Li, Hongqi Wang, Xian Sun
2020 arXiv   pre-print
In this paper, we propose a novel model named Oriented Objects Detection Network O^2-DNet to detect oriented objects by predicting a pair of middle lines inside each target.  ...  O^2-DNet is an one-stage, anchor-free and NMS-free model.  ...  In this paper, we propose a novel anchor-free and NMS-free model named O 2 -DNet as shown in Fig. 2 to detect oriented objects by a pair of middle lines.  ... 
arXiv:1912.10694v3 fatcat:2sgxna5h7zddhhfopyd4y6kfzu

Shape-Adaptive Selection and Measurement for Oriented Object Detection

Liping Hou, Ke Lu, Jian Xue, Yuqiu Li
2022 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
In this paper, we propose novel flexible shape-adaptive selection (SA-S) and shape-adaptive measurement (SA-M) strategies for oriented object detection, which comprise an SA-S strategy for sample selection  ...  The development of detection methods for oriented object detection remains a challenging task. A considerable obstacle is the wide variation in the shape (e.g., aspect ratio) of objects.  ...  There are five types of detection methods for oriented object detection: (1) generating oriented region proposals directly (Azimi et al. 2018; Ding et al. 2019 ); (2) regressing the angle parameters based  ... 
doi:10.1609/aaai.v36i1.19975 fatcat:swrtcmdrxfchppdrdxck4chxha

SAFDet: A Semi-Anchor-Free Detector for Effective Detection of Oriented Objects in Aerial Images

Zhenyu Fang, Jinchang Ren, He Sun, Stephen Marshall, Junwei Han, Huimin Zhao
2020 Remote Sensing  
To this end, we propose a semi-anchor-free detector (SAFDet) for object detection in aerial images, where a rotation-anchor-free-branch (RAFB) is used to enhance the foreground features via precisely regressing  ...  An oriented bounding box (OBB) is preferable over a horizontal bounding box (HBB) in accurate object detection.  ...  Anchor-Free Object Detection The anchor-free strategy was originally designed for general object detection [23] [24] [25] [26] .  ... 
doi:10.3390/rs12193225 fatcat:6n33kib4dvg3lpgq7d4e3ay4xm

TricubeNet: 2D Kernel-Based Object Representation for Weakly-Occluded Oriented Object Detection [article]

Beomyoung Kim, Janghyeon Lee, Sihaeng Lee, Doyeon Kim, Junmo Kim
2021 arXiv   pre-print
We present a novel approach for oriented object detection, named TricubeNet, which localizes oriented objects using visual cues (i.e., heatmap) instead of oriented box offsets regression.  ...  To demonstrate the effectiveness of our TricubeNet, we experiment on various tasks for weakly-occluded oriented object detection: detection in an aerial image, densely packed object image, and text image  ...  Anchor-free object detectors have recently been proposed and eliminate the anchor box in the network design.  ... 
arXiv:2104.11435v2 fatcat:m45tsovg4rcpbew3gvwlatxdmy

From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Network [article]

Shaoshuai Shi, Zhe Wang, Jianping Shi, Xiaogang Wang, Hongsheng Li
2020 arXiv   pre-print
Firstly, the part-aware stage for the first time fully utilizes free-of-charge part supervisions derived from 3D ground-truth boxes to simultaneously predict high quality 3D proposals and accurate intra-object  ...  Our Part-A^2 net outperforms all existing 3D detection methods and achieves new state-of-the-art on KITTI 3D object detection dataset by utilizing only the LiDAR point cloud data.  ...  Since our proposed bottom-up proposal generation strategy is anchor-free, it does not have an initial value for box orientation.  ... 
arXiv:1907.03670v3 fatcat:ungdwipuwzhprpd2ymgcir3cdy

ArbiText: Arbitrary-Oriented Text Detection in Unconstrained Scene [article]

Daitao Xing, Zichen Li, Xin Chen, Yi Fang
2017 arXiv   pre-print
In this paper, we propose a novel method, namely Arbitrary-oriented Text (or ArbText for short) detector, for efficient text detection in unconstrained natural scene images.  ...  Arbitrary-oriented text detection in the wild is a very challenging task, due to the aspect ratio, scale, orientation, and illumination variations.  ...  Conclusion We have presented ArbiText, a novel, proposal-free object detection method that can be utilized to detect both arbitrary-oriented texts and generic objects simultaneously.  ... 
arXiv:1711.11249v1 fatcat:jcqx4vgo7jfqjd4ngb6wnzuhru

REGNet: REgion-based Grasp Network for End-to-end Grasp Detection in Point Clouds [article]

Binglei Zhao, Hanbo Zhang, Xuguang Lan, Haoyu Wang, Zhiqiang Tian, Nanning Zheng
2021 arXiv   pre-print
To further improve the performance, we propose a grasp anchor mechanism, in which grasp anchors with assigned gripper orientations are introduced to generate grasp proposals.  ...  The main problem is to generate the optimal grasp of novel objects from partial noisy observations.  ...  Grasp Detection Existing grasp detection methods are generally divided into two categories: model-based and model-free.  ... 
arXiv:2002.12647v3 fatcat:n5ndgmgbfjcwnit2fi3fof6uz4

Self-Adaptive Aspect Ratio Anchor for Oriented Object Detection in Remote Sensing Images

Jie-Bo Hou, Xiaobin Zhu, Xu-Cheng Yin
2021 Remote Sensing  
Generally, existing object detection methods ignore exploring the distribution character of aspect ratios for improving performance in remote sensing tasks.  ...  In this paper, we propose a novel Self-Adaptive Aspect Ratio Anchor (SARA) to explicitly explore aspect ratio variations of objects in remote sensing images.  ...  The author would like to express their appreciation to the developers of pytorch and S 2 A-Net for their open source code.  ... 
doi:10.3390/rs13071318 fatcat:6k2fgzmf3nfobceif2vhjej7oq

Point RCNN: An Angle-Free Framework for Rotated Object Detection

Qiang Zhou, Chaohui Yu
2022 Remote Sensing  
To tackle this problem, we propose a purely angle-free framework for rotated object detection, called Point RCNN.  ...  Rotated object detection in aerial images is still challenging due to arbitrary orientations, large scale and aspect ratio variations, and extreme density of objects.  ...  Conclusions In this study, we revisited rotated object detection and proposed a purely angle-free framework for rotated object detection, named Point RCNN, which mainly consists of a PointRPN for generating  ... 
doi:10.3390/rs14112605 fatcat:4tx2ykxiineuhcpqmn4w7vuq6e
« Previous Showing results 1 — 15 out of 48,704 results