Filters








169,271 Hits in 10.7 sec

Bottom-up Object Detection by Grouping Extreme and Center Points [article]

Xingyi Zhou, Jiacheng Zhuo, Philipp Krähenbühl
2019 arXiv   pre-print
We detect four extreme points (top-most, left-most, bottom-most, right-most) and one center point of objects using a standard keypoint estimation network.  ...  With the advent of deep learning, object detection drifted from a bottom-up to a top-down recognition problem.  ...  Acknowledgement We thank Chao-Yuan Wu, Dian Chen, and Chia-Wen Cheng for helpful feedback.  ... 
arXiv:1901.08043v3 fatcat:tx55bisflvfgpoxppcjrj4h6du

Bottom-Up Object Detection by Grouping Extreme and Center Points

Xingyi Zhou, Jiacheng Zhuo, Philipp Krahenbuhl
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We detect four extreme points (top-most, leftmost, bottom-most, right-most) and one center point of objects using a standard keypoint estimation network.  ...  With the advent of deep learning, object detection drifted from a bottom-up to a top-down recognition problem.  ...  Acknowledgement We thank Chao-Yuan Wu, Dian Chen, and Chia-Wen Cheng for helpful feedback.  ... 
doi:10.1109/cvpr.2019.00094 dblp:conf/cvpr/ZhouZK19 fatcat:vaykxd4rtvbijm47glz3amdyme

HPRNet: Hierarchical Point Regression for Whole-Body Human Pose Estimation [article]

Nermin Samet, Emre Akbas
2021 arXiv   pre-print
In this paper, we present a new bottom-up one-stage method for whole-body pose estimation, which we call "hierarchical point regression," or HPRNet for short.  ...  On the COCO WholeBody dataset, HPRNet significantly outperforms all previous bottom-up methods on the keypoint detection of all whole-body parts (i.e. body, foot, face and hand); it also achieves state-of-the-art  ...  Our method is based on the center-point based bottom-up object detection methods [60, 46, 13, 31] . These methods can easily be extended to the keypoint estimation task [60, 47] .  ... 
arXiv:2106.04269v2 fatcat:vcmdn6mr5ffyligr3cnla2cjta

RECIST-Net: Lesion detection via grouping keypoints on RECIST-based annotation [article]

Cong Xie, Shilei Cao, Dong Wei, Hongyu Zhou, Kai Ma, Xianli Zhang, Buyue Qian, Liansheng Wang, Yefeng Zheng
2021 arXiv   pre-print
(RECIST) diameters, we propose RECIST-Net, a new approach to lesion detection in which the four extreme points and center point of the RECIST diameters are detected.  ...  By detecting a lesion as keypoints, we provide a more conceptually straightforward formulation for detection, and overcome several drawbacks (e.g., requiring extensive effort in designing data-appropriate  ...  The RECIST-Net detected four extreme points (i.e., top-most, leftmost, bottom-most, right-most) and one center point of a lesion in a keypoint detection way.  ... 
arXiv:2107.08715v1 fatcat:yqgxrowtjbh7ze7eby7fwyubfe

Semantic Segmentation with Oblique Convolution for Object Detection

Yun Lin, Xiaogang Sun, Zhixuan Xie, Jiaqi Yi, Yong Zhong
2020 IEEE Access  
Instead of anchors, we use pixel classification inspired by Semantic Segmentation to get the local extreme points of the four boundaries of an object and then the boundary positions.  ...  We calculate the possibility whether every pixel in the image is the extreme point by hourglass network.  ...  ACKNOWLEDGMENT All the machines used in the experiment were from Xinda of China Banknote Printing and Minting.  ... 
doi:10.1109/access.2020.2971058 fatcat:ztlyszsm2ndbznob6r4wfkhr3e

Spatial-Temporal Distribution Analysis of Industrial Heat Sources in the US with Geocoded, Tree-Based, Large-Scale Clustering

Yan Ma, Caihong Ma, Peng Liu, Jin Yang, Yuzhu Wang, Yueqin Zhu, Xiaoping Du
2020 Remote Sensing  
Although the object-oriented approach, especially the data clustering-based approach, has guided a novel method of detection, it is still limited by the costly computation and storage resources.  ...  Furthermore, in order to avoid the blindness caused by random cluster center initialization, the time series VIIRS hotspots data are spatially pre-grouped into GeoSOT-encoded grid tasks which are also  ...  tree in a bottom-up way.  ... 
doi:10.3390/rs12183069 doaj:418ec69a815e42fbb4b4675c67e40bc2 fatcat:krkzw3q5zzbjjnt2jld3axrrfu

PointNu-Net: Simultaneous Multi-tissue Histology Nuclei Segmentation and Classification in the Clinical Wild [article]

Kai Yao and Kaizhu Huang and Jie Sun and Amir Hussain and Curran Jude
2021 arXiv   pre-print
We address the detection and classification of each nuclei as a novel semantic keypoint estimation problem to determine the center point of each nuclei.  ...  By decoupling two simultaneous challenging tasks, our method can benefit from class-aware detection and class-agnostic segmentation, thus leading to a significant performance boost.  ...  CenterNet [19] located objects by the center points and regressed the corresponding size.  ... 
arXiv:2111.01557v1 fatcat:ruwivk7uobcdzjxm2rbzk2rqde

SaccadeNet: A Fast and Accurate Object Detector [article]

Shiyi Lan, Zhou Ren, Yi Wu, Larry S. Davis, Gang Hua
2020 arXiv   pre-print
Object detection is an essential step towards holistic scene understanding. Most existing object detection algorithms attend to certain object areas once and then predict the object locations.  ...  It contains four main modules, the , the , the , and the , which allows it to attend to different informative object keypoints, and predict object locations from coarse to fine.  ...  We introduce SaccadeNet, a fast and accurate object detection algorithm.  ... 
arXiv:2003.12125v1 fatcat:tkbrncvcwves7mer2zugeh5bjy

EOLO: Deep Machine Learning Algorithm for Embedded Object Segmentation that Only Looks Once

Longfei Zeng, Lakehead University, Smart Health FabLab, Department of Computer Science, Canada, Sabah Mohammed, Lakehead University, Smart Health FabLab, Department of Computer Science, Canada
2020 International Journal of Multimedia and Ubiquitous Engineering  
For the first time, we show the different comprehension of instance segmentation in recent methods, in terms of both up-bottoms, down-ups, and direct-predict paradigms.  ...  Our method, refer as EOLO, reformulates the instance segmentation problem as predicting semantic segmentation and distinguishing overlapping objects problem, through instance center classification and  ...  While bottom-up approaches generate instance masks by grouping the pixels into a set of candidate masks in an image and embed, cluster and assemble them.  ... 
doi:10.21742/ijmue.2020.15.1.04 fatcat:lu65v6ti5rdfvb27ge3mgtcncm

RepPoints: Point Set Representation for Object Detection [article]

Ze Yang and Shaohui Liu and Han Hu and Liwei Wang and Stephen Lin
2019 arXiv   pre-print
In this paper, we present RepPoints (representative points), a new finer representation of objects as a set of sample points useful for both localization and recognition.  ...  We show that an anchor-free object detector based on RepPoints can be as effective as the state-of-the-art anchor-based detection methods, with 46.5 AP and 67.4 AP_50 on the COCO test-dev detection benchmark  ...  These bottom-up representations identify individual points (e.g., bounding box corners or object extremities) and rely on handcrafted clustering to group them into object models.  ... 
arXiv:1904.11490v2 fatcat:zlks3bzjnfgy5bfh3imnc4aie4

Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation

Bowen Cheng, Maxwell D. Collins, Yukun Zhu, Ting Liu, Thomas S. Huang, Hartwig Adam, Liang-Chieh Chen
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
In this work, we introduce Panoptic-DeepLab, a simple, strong, and fast system for panoptic segmentation, aiming to establish a solid baseline for bottom-up methods that can achieve comparable performance  ...  For the first time, we demonstrate a bottom-up approach could deliver state-of-the-art results on panoptic segmentation.  ...  CornerNet [38] detects objects by predicting paired corners and group corners based on [56] . ExtremeNet [79] groups 'extreme points' [57] according to the relation to a center point.  ... 
doi:10.1109/cvpr42600.2020.01249 dblp:conf/cvpr/ChengCZ0HAC20 fatcat:ejjxoxzuhjdmllbg7qgiadn5a4

HoughNet: Integrating near and long-range evidence for bottom-up object detection [article]

Nermin Samet, Samet Hicsonmez, Emre Akbas
2020 arXiv   pre-print
This paper presents HoughNet, a one-stage, anchor-free, voting-based, bottom-up object detection method.  ...  On the COCO dataset, HoughNet's best model achieves 46.4 AP (and 65.1 AP_50), performing on par with the state-of-the-art in bottom-up object detection and outperforming most major one-stage and two-stage  ...  Acknowledgments This work was supported by the Scientific and Technological Research Council of Turkey (TÜBİTAK) through the project titled "Object Detection in Videos with Deep Neural Networks" (grant  ... 
arXiv:2007.02355v3 fatcat:qpvx6zow5fhztlptdsz4ld6jha

SaccadeNet: A Fast and Accurate Object Detector

Shiyi Lan, Zhou Ren, Yi Wu, Larry S. Davis, Gang Hua
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Inspired by such mechanism, we propose a fast and accurate object detector called SaccadeNet.  ...  Object detection is an essential step towards holistic scene understanding. Most existing object detection algorithms attend to certain object areas once and then predict the object locations.  ...  Acknowledgement Gang Hua was supported partly by National Key R&D Program of China Grant 2018AAA0101400 and NSFC Grant 61629301. We deeply appreciate the help of Xingyi Zhou and Zuxuan Wu.  ... 
doi:10.1109/cvpr42600.2020.01041 dblp:conf/cvpr/LanRWD020 fatcat:kfocvkodgzanxp5jyqd3vzjefu

DIAMONDNET: SHIP DETECTION IN REMOTE SENSING IMAGES BY EXTRACTING AND CLUSTERING KEYPOINTS IN A DIAMOND

Z. Zhu, W. Diao, K. Chen, L. Zhao, Z. Yan, W. Zhang, G. Xu, X. Sun
2020 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
We use a convolutional neural network to detect ship keypoints and then cluster the keypoints into groups, where each group is composed of keypoints belonging to the same ship.  ...  In this work, we come up with a novel ship detection framework based on the keypoint extraction technique.  ...  On the contrary, the bottom-up method first extracts the keypoints and then clusters the keypoints into groups so as to find the keypoints belonging to a single object (Cao et al., 2017) .  ... 
doi:10.5194/isprs-annals-v-2-2020-625-2020 fatcat:hlew3razmvdgtflvosbv4rxycq

Traffic Sign Detection and Recognition Using Novel Center-Point Estimation and Local Features

Lijing Wei, Cheng Xu, Siqi Li, Xiaohan Tu
2020 IEEE Access  
INDEX TERMS Traffic sign detection, multi-scale, center-point estimation, local features.  ...  Specially, the prediction is made on the fuse feature map and based on the novel center-point estimation.  ...  CenterNet [20] simply detects the center point of each object which avoids the time-consuming and error-prone grouping step and performs better than Cor-nerNet.  ... 
doi:10.1109/access.2020.2991195 fatcat:bgjdr4v6qzhwpla24wgkgermzq
« Previous Showing results 1 — 15 out of 169,271 results