面向算力受限边缘环境的双分支多尺度感知人脸检测网络

戚琦; 马迎新; 王敬宇; 孙海峰; 廖建新

doi:10.11959/j.issn.1000-436x.2020177

您当前的位置：

首页 >

文章列表页 >

面向算力受限边缘环境的双分支多尺度感知人脸检测网络

学术论文 | 更新时间：2024-06-05

- 面向算力受限边缘环境的双分支多尺度感知人脸检测网络
- Multi-scale aware dual path network for face detection in resource-constrained edge computing environment
- 通信学报 2020年41卷第8期页码：165-174
- 作者机构：
  
  北京邮电大学网络与交换国家重点实验室，北京 100876
- 作者简介：
  
  [ "戚琦（1982– ），女，河北廊坊人，博士，北京邮电大学副教授、博士生导师，主要研究方向为智能边缘计算、轻量级神经网络、业务网络智能化等" ]
  [ "马迎新（1996- ），男，山西临汾人，北京邮电大学硕士生，主要研究方向为深度学习、计算机视觉、人脸检测与识别" ]
  [ "王敬宇（1978- ），男，吉林长春人，博士，北京邮电大学教授、博士生导师，主要研究方向为智能网络、人工智能、计算机视觉、深度学习、多媒体通信等" ]
  [ "孙海峰（1989– ），男，天津人，博士，北京邮电大学讲师、硕士生导师，主要研究方向为人工智能、机器视觉、自然语言处理、深度学习等" ]
  [ "廖建新（1965– ），男，四川宜宾人，博士，北京邮电大学“长江学者”特聘教授、博士生导师，主要研究方向为移动通信网络、业务网络化、人工智能、多媒体业务等" ]
- 基金信息：
  
  国家重点研发计划基金资助项目(2018YFB1800502);国家自然科学基金资助项目(61671079);国家自然科学基金资助项目(61771068);北京市自然科学基金资助项目(4182041)
- DOI：10.11959/j.issn.1000-436x.2020177
  中图分类号： TP391.41
- 网络出版日期：2020-08，
  
  纸质出版日期：2020-08-25
- 稿件说明：
移动端阅览
戚琦, 马迎新, 王敬宇, 等. 面向算力受限边缘环境的双分支多尺度感知人脸检测网络[J]. 通信学报, 2020,41(8):165-174.

Qi QI, Yingxin MA, Jingyu WANG, et al. Multi-scale aware dual path network for face detection in resource-constrained edge computing environment[J]. Journal on communications, 2020, 41(8): 165-174.
戚琦, 马迎新, 王敬宇, 等. 面向算力受限边缘环境的双分支多尺度感知人脸检测网络[J]. 通信学报, 2020,41(8):165-174. DOI： 10.11959/j.issn.1000-436x.2020177.

Qi QI, Yingxin MA, Jingyu WANG, et al. Multi-scale aware dual path network for face detection in resource-constrained edge computing environment[J]. Journal on communications, 2020, 41(8): 165-174. DOI： 10.11959/j.issn.1000-436x.2020177.

摘要

针对边缘算力受限，难以部署复杂结构的人脸检测深度神经网络的问题，为减少资源消耗，并保证人脸在多尺度变化、遮挡、模糊、光照等复杂场景下的检测精度，提出了多尺度感知的轻量化人脸检测算法。采用改进的人脸残差神经网络作为特征提取网络，并提出双分支浅层特征提取模块，并行分支理解图像多尺度信息，进而由深浅特征融合模块将底层图像信息与高层语义特征融合，配合多尺度感知的训练策略监督多分支学习差异化特征。实验结果表明，所提算法可有效提取多样化的特征，在保持模型高效性和低推理时延的同时，有效提升了算法的精度和稳健性。

Abstract

Aiming at the problem that face detectors with complex deep neural structures are difficult to deploy in the resource-constrained edge computing environment

to reduce the resource consumption while maintain the accuracy in complex scenes such as multi-scale face changes

occlusion

blur

and illumination

SDPN(multi-scale aware dual path network) for face detection was proposed.The Face-ResNet (face residual neural network) was improved

and a dual path shallow feature extractor was used to understand the multi-scale information of the image through parallel branches.Then the deep and shallow feature fusion module

a combination of the underlying image information and the high-level semantic feature

was used in conjunction with the multi-scale awareness training strategy to supervise the multi-branch learning discriminating features.The experimental results show that SDPN can extract more diversified features

which effectively improve the accuracy and robustness of face detection while maintaining the efficiency of the model and low inference delay.

关键词

Keywords

references

HU P , RAMANAN D . Finding tiny faces [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . Piscataway:IEEE Press , 2017 : 951 - 959 .

CHI C , ZHANG S , XING J , et al . Selective refinement network for high performance face detection [C ] // Proceedings of the AAAI Conference on Artificial Intelligence . Palo Alto:AAAI Press , 2019 : 8231 - 8238 .

LI J , WANG Y , WANG C , et al . DSFD:dual shot face detector [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . Piscataway:IEEE Press , 2019 : 5060 - 5069 .

NAJIBI M , SAMANGOUEI P , CHELLAPPA R , et al . SSH:single stage headless face detector [C ] // Proceedings of the IEEE International Conference on Computer Vision . Piscataway:IEEE Press , 2017 : 4875 - 4884 .

TANG X , DU D K , HE Z , et al . Pyramidbox:a contextassisted single shot face detector [C ] // Proceedings of the European Conference on Computer Vision . Berlin:Springer , 2018 : 797 - 813 .

HE Y , XU D , WU L , et al . LFFd:a light and fast face detector for edge devices [J ] . arXiv Preprint,arXiv:1904.10633 , 2019

ZHANG S , ZHU X , LEI Z , et al . Faceboxes:a CPU real-time face detector with high accuracy [C ] // 2017 IEEE International Joint Conference on Biometrics . Piscataway:IEEE Press , 2017 : 1 - 9 .

YANG S , LUO P , LOY C C , et al . Wider face:a face detection benchmark [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . Piscataway:IEEE Press , 2016 : 5525 - 5533 .

DENG J , DONG W , SOCHER R , et al . ImageNet:a largescale hierarchical image database [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern . Piscataway:IEEE Press , 2009 : 248 - 255 .

REN S , HE K , GIRSHICK R , et al . Faster R-CNN:towards real-time object detection with region proposal networks [C ] // Neural Information Processing Systems . Massachusetts:MIT Press , 2015 : 91 - 99 .

LIN T Y , DOLLAR P , GIRSHICK R , et al . Feature pyramid networks for object detection [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern . Piscataway:IEEE Press , 2017 : 2117 - 2125 .

HUANG L , YANG Y , DENG Y , et al . Densebox:unifying landmark localization with end to end object detection [J ] . arXiv Preprint,arXiv:1509.04874 , 2015

ZHANG K , ZHANG Z , LI Z , et al . Joint face detection and alignment using multitask cascaded convolutional networks [J ] . IEEE Signal Processing Letters , 2016 , 23 ( 10 ): 1499 - 1503 .

JIANG H , LEARNED-MILLER E , . Face detection with the faster R-CNN [C ] // 2017 12th IEEE International Conference on Automatic Face ＆ Gesture Recognition . Piscataway:IEEE Press , 2017 : 650 - 657 .

LIU W , ANGUELOV D , ERHAN D , et al . SSD:single shot multibox detector [C ] // Proceedings of the European Conference on Computer Vision . Berlin:Springer , 2016 : 21 - 37 .

REDMON J , DIVVALA S , GIRSHICK R , et al . You only look once:unified,real-time object detection [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . Piscataway:IEEE Press , 2016 : 779 - 788 .

ZHANG S , ZHU X , LEI Z , et al . S3FD:single shot scaleinvariant face detector [C ] // Proceedings of the IEEE International Conference on Computer Vision . Piscataway:IEEE Press , 2017 : 192 - 201 .

LAW H , DENG J . Cornernet:detecting objects as paired keypoints [C ] // Proceedings of the European Conference on Computer Vision . Berlin:Springer , 2018 : 734 - 750 .

DUAN , KAIWEN , , et al . Centernet:keypoint triplets for object detection [C ] // Proceedings of the IEEE International Conference on Computer Vision . Piscataway:IEEE Press , 2019 : 6569 - 6578 .

LIU W , LIAO S , REN W , et al . High-level semantic feature detection:a new perspective for pedestrian detection [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . Piscataway:IEEE Press , 2019 : 5187 - 5196 .

HE K , ZHANG X , REN S , et al . Deep residual learning for image recognition [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . Piscataway:IEEE Press , 2016 : 770 - 778 .

ODENA A , DUMOULIN V , OLAH C . Deconvolution and checkerboard artifacts [J ] . Distill , 2016 , 1 ( 10 ):e3.

SINGH B , DAVIS L S . An analysis of scale invariance in object detection SNIP [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . Piscataway:IEEE Press , 2018 : 3578 - 3587 .

LIN T , GOYAL P , GIRSHICK R , et al . Focal loss for dense object detection [C ] // Proceedings of the IEEE International Conference on Computer vision . Piscataway:IEEE Press , 2017 : 2980 - 2988 .

浏览量

566

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于非平稳信号组合分析的故障诊断方法

超大规模太赫兹系统深度学习信道估计算法

基于多粒度融合和跨尺度感知的跨模态行人重识别

基于机器学习的加密流量分类研究综述

基于深度学习的SDN异常流量分布式检测方法