Filters








10,214 Hits in 5.6 sec

An Improved Faster R-CNN for High-Speed Railway Dropper Detection

Qifan Guo, Lei Liu, Wenjuan Xu, Yansheng Gong, Xuewu Zhang, Wenfeng Jing
2020 IEEE Access  
First, a balanced attention feature pyramid network (BA-FPN) is used to predict the detection anchor.  ...  INDEX TERMS Dropper detection, feature fusion, improved Faster R-CNN, attention mechanism. 105622 This work is licensed under a Creative Commons Attribution 4.0 License.  ...  CONCLUSION This paper proposes an improved Faster R-CNN for OCS dropper detection, including the balanced attention feature pyramid network (BA-FPN) and center-point rectangle loss (CR loss).  ... 
doi:10.1109/access.2020.3000506 fatcat:zwbmpjxsrfei5ipdhe2rho6m6a

MSB R-CNN: A Multi-Stage Balanced Defect Detection Network

Zhihua Xu, Shangwei Lan, Zhijing Yang, Jiangzhong Cao, Zongze Wu, Yongqiang Cheng
2021 Electronics  
Deep learning networks are applied for defect detection, among which Cascade R-CNN is a multi-stage object detection network and is state of the art in terms of accuracy and efficiency.  ...  To address the above challenges, this paper proposes a multi-stage balanced R-CNN (MSB R-CNN) for defect detection based on Cascade R-CNN.  ...  The Proposed MSB R-CNN MSB R-CNN is an object detection network designed for defect detection. It can better balance the learning of the network and effectively improve the detection accuracy.  ... 
doi:10.3390/electronics10161924 fatcat:es7h4ltxj5bojpifedgkncbapy

A New Deep Learning Network for Automatic Bridge Detection from SAR Images Based on Balanced and Attention Mechanism

Chen, Weng, Xing, Pan, Yuan, Xing, Zhang
2020 Remote Sensing  
In addition, intersection over union (IOU) balanced sampling and balanced L1 loss functions are introduced for optimal training of the classification and regression network.  ...  It mainly includes three parts, the attention and balanced feature pyramid (ABFP) network, the region proposal network (RPN), and the classification and regression.  ...  Therefore, objects detection and classification turn out to be more challenging for SAR images.  ... 
doi:10.3390/rs12030441 fatcat:mllzgqmqu5b2ngp4fxkblfbs64

Libra R-CNN: Towards Balanced Learning for Object Detection [article]

Jiangmiao Pang, Kai Chen, Jianping Shi, Huajun Feng, Wanli Ouyang, Dahua Lin
2019 arXiv   pre-print
To mitigate the adverse effects caused thereby, we propose Libra R-CNN, a simple but effective framework towards balanced learning for object detection.  ...  Benefitted from the overall balanced design, Libra R-CNN significantly improves the detection performance.  ...  Balanced learning for object detection.  ... 
arXiv:1904.02701v1 fatcat:pp3fn57dijegxas2b2bh6t5ppq

Libra R-CNN: Towards Balanced Learning for Object Detection

Jiangmiao Pang, Kai Chen, Jianping Shi, Huajun Feng, Wanli Ouyang, Dahua Lin
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
To mitigate the adverse effects caused thereby, we propose Libra R-CNN, a simple but effective framework towards balanced learning for object detection.  ...  Benefitted from the overall balanced design, Libra R-CNN significantly improves the detection performance.  ...  Balanced learning for object detection.  ... 
doi:10.1109/cvpr.2019.00091 dblp:conf/cvpr/PangCSFOL19 fatcat:dytsvhfa2zffhlmcy3sazqkp4u

ThunderNet: Towards Real-time Generic Object Detection [article]

Zheng Qin, Zeming Li, Zhaoning Zhang, Yiping Bao, Gang Yu, Yuxing Peng, Jian Sun
2022 arXiv   pre-print
To generate more discriminative feature representation, we design two efficient architecture blocks, Context Enhancement Module and Spatial Attention Module.  ...  In the backbone part, we analyze the drawbacks in previous lightweight backbones and present a lightweight backbone designed for object detection.  ...  This work is sponsored in part by the National Key R&D Program of China (2018YFB2101100, 2017YFA0700800).  ... 
arXiv:1903.11752v3 fatcat:rjhdbocbhzgkph6fcg3dezsnei

Joint Object Contour Points and Semantics for Instance Segmentation [article]

Wenchao Zhang, Chong Fu, Mai Zhu
2021 arXiv   pre-print
As a consequence, the model will be more sensitive to the edges of the object and can capture more geometric features.  ...  Inspired by the human annotation process when making instance segmentation datasets, in this paper, we propose Mask Point R-CNN aiming at promoting the neural network's attention to the object boundary  ...  In other words, we add a new keypoint detection auxiliary task to Mask R-CNN, which can enhance the model's attention to the object boundary by detecting the contour of the object.  ... 
arXiv:2008.00460v3 fatcat:6ufmuob7oja5zl6xgcgztw46cm

A Cascaded R-CNN with Multiscale Attention and Imbalanced Samples for Traffic Sign Detection

Jianming Zhang, Zhipeng Xie, Juan Sun, Xin Zou, Jin Wang
2020 IEEE Access  
Therefore, to solve the undetection and false detection, we first propose a cascaded R-CNN to obtain the multiscale features in pyramids.  ...  INDEX TERMS Traffic sign detection, convolutional neural network, attention, object detection, Multiscale.  ...  Except for the indicators given in Table 1 , the average execution time of our model is 7.6 fps,while the average execution time in Faster R-CNN, Mask R-CNN, and Cascade R-CNN is 2.26 fps, 5.7 fps, 7  ... 
doi:10.1109/access.2020.2972338 fatcat:oiyel6vctfa3jblwt3dvu6lnva

Object Detection Network Based on Feature Fusion and Attention Mechanism

Ying Zhang, Yimin Chen, Chen Huang, Mingke Gao
2019 Future Internet  
Also, the attention mechanism was applied to our object detection network, AF R-CNN (attention mechanism and convolution feature fusion based object detection), to enhance the impact of significant features  ...  Our AF R-CNN achieves an object detection accuracy of 75.9% on PASCAL VOC 2007, six points higher than Faster R-CNN.  ...  Middle: The object detection of the Faster R-CNN. Right: The object detection of AF R-CNN. We found a few more representative pictures.  ... 
doi:10.3390/fi11010009 fatcat:dqaciv3llbgcrbmx4s7vmkiapm

Few-Shot Object Detection on Remote Sensing Images via Shared Attention Module and Balanced Fine-Tuning Strategy

Xu Huang, Bokun He, Ming Tong, Dingwen Wang, Chu He
2021 Remote Sensing  
Few-shot object detection is a recently emerging branch in the field of computer vision. Recent research studies have proposed several effective methods for object detection with few samples.  ...  characteristics of remote sensing images to enhance the few-shot detection performance.  ...  of interest mAP Mean average precision SAM Shared attention module BFS Balanced fine tuning strategy  ... 
doi:10.3390/rs13193816 fatcat:msfyq3bukjgbhkkjfpone5ynfu

IoU-balanced Loss Functions for Single-stage Object Detection [article]

Shengkai Wu, Jinrong Yang, Xinggang Wang, Xiaoping Li
2020 arXiv   pre-print
The IoU-balanced classification loss pays more attention to positive examples with high IoU and can enhance the correlation between classification and localization tasks.  ...  Extensive experiments on challenging public datasets such as MS COCO, PASCAL VOC and Cityscapes demonstrate that both IoU-balanced losses can bring substantial improvement for the popular single-stage  ...  Xuzhan Chen for their advice and language help.  ... 
arXiv:1908.05641v2 fatcat:bu6clylgondbzjetk2q5h22nha

Modified Deep Reinforcement Learning with Efficient Convolution Feature for Small Target Detection in VHR Remote Sensing Imagery

Shuai Liu, Jialan Tang
2021 ISPRS International Journal of Geo-Information  
It also can increase the effectiveness of object locations and classifications for small targets.  ...  By this, the attention network can effectively enhance the ability to extract small target features and suppressing useless features.  ...  Its success can be contributed to the balance between the classification and localization for object detection.  ... 
doi:10.3390/ijgi10030170 fatcat:7657phniufey7nfxyte3d2gkme

Feature Enhancement Network for Object Detection in Optical Remote Sensing Images

Gong Cheng, Chunbo Lang, Maoxiong Wu, Xingxing Xie, Xiwen Yao, Junwei Han
2021 Journal of Remote Sensing  
In this paper, we propose a novel Feature Enhancement Network (FENet) for object detection in optical remote sensing images, which consists of a Dual Attention Feature Enhancement (DAFE) module and a Context  ...  We achieve our proposed FENet by unifying DAFE and CFE into the framework of Faster R-CNN.  ...  Inspired by Mask R-CNN, [40] proposed a refine FPN and multilayer attention network for oriented object detection of remote sensing images. Proposed Method Review of Faster R-CNN.  ... 
doi:10.34133/2021/9805389 fatcat:y2hxasy5mjcqjn7y24glaknxem

Intelligent Vision-Enabled Detection of Water-Surface Targets for Video Surveillance in Maritime Transportation

Yongqi Guo, Yuxu Lu, Yu Guo, Ryan Wen Liu, Kwok Tai Chui, Xinqiang Chen
2021 Journal of Advanced Transportation  
The timely, automatic, and accurate detection of water-surface targets has received significant attention in intelligent vision-enabled maritime transportation systems.  ...  The reliable detection results are also beneficial for water quality monitoring in practical applications.  ...  For more implementation details on Faster R-CNN, please refer to [10] and references therein.  ... 
doi:10.1155/2021/9470895 fatcat:rcznedvb7fgurkma43wuwef3za

Towards Balanced Learning for Instance Recognition

Jiangmiao Pang, Kai Chen, Qi Li, Zhihai Xu, Huajun Feng, Jianping Shi, Wanli Ouyang, Dahua Lin
2021 International Journal of Computer Vision  
To mitigate the adverse effects caused thereby, we propose Libra R-CNN, a simple yet effective framework towards balanced learning for instance recognition.  ...  It integrates IoU-balanced sampling, balanced feature pyramid, and objective re-weighting, respectively for reducing the imbalance at sample, feature, and objective level.  ...  [20] that diagnoses errors in object detection, we further analyze the errors both in Faster R-CNN and Libra R-CNN.  ... 
doi:10.1007/s11263-021-01434-2 fatcat:lvblzs5cnzfz7daf2exgx6u3pq
« Previous Showing results 1 — 15 out of 10,214 results