学术信息网 西电导航 关于 使用说明 搜索 系统首页 登录 控制面板 收藏 谢雪梅的留言板
基本信息

谢雪梅 教授    硕导/博导

博士学科:电路与系统(080902)080902)
 ((()()()dssfd1

硕士学科:电路与系统(080902)

工作单位:人工智能学院

联系方式

通信地址:西安电子科技大学134信箱

电子邮箱:xmxie@mail.xidian.edu.cn

办公电话:029-88204453

办公地点:北校区主楼II-134

ORCID: https://orcid.org/0000-0001-7857-0845

个人简介

谢雪梅    教授,博导

目前感兴趣的研究方向:

强化学习和深度学习 (Reinforcement Learning & Deep Learning)

图像重构和压缩感知 (Image Reconstruction & Compressive Sensing)

 

人体异常行为及目标检测 (Human Action Recognition & Object Detection)

视频与图像处理 (Video & Image Processing)

 

脑启发式计算 (Brain-inspired Computation)

 

主持完成国家自然科学基金面上和重点项目、863项目、博士点基金、“新世纪优秀人才”基金等。

主持:自然基金重点项目,面向弱小目标的机器学习与智能认知方法(61836008)

 

 

  

教育及工作经历

1998-2004   香港大学       博士

1991-1994   西安电子科技大学  硕士

2007 至今   西安电子科技大学  教授

2005-2007   西安电子科技大学  副教授

2004.3-2004.7 香港大学       助理研究员(R.A.) 

1986-1991 & 1994-1998 & 2004 至今 西安电子科技大学 教师 

主要研究方向

1. Deep Reinforcement Learning

Papers:

[1] Xuemei Xie, et al. Skeleton action recognition based on deep reinforcement learning, to be submitted.

[2] Xuemei Xie, et al. Attention based measurement network for compressive reinforcement sensing, to be submitted.

 

2. Image Reconstruction & Compressive Sensing (CS)

项目:面向大数据的预期形状导向压缩感知成像的重构

Papers:

[1] Jiang Du, Xuemei Xie, Chenye Wang, Guangming Shi, Xun Xu, Yuxiang Wang, Fully convolutional measurement network for compressive sensing image reconstruction [J]. Neurocomputing, 2018.

[2] Jiang Du, Xuemei Xie, Chenye Wang, Guangming Shi, Color image reconstruction with perceptual compressive sensing, IEEE International Conference on Pattern Recognition (ICPR), August 2018.

[3] Jiang Du, Xuemei Xie, Chenye Wang, Guangming Shi, Perceptual Compressive Sensing, The First Chinese Conference on Pattern Recognition and Computer Vision (PRCV), November 2018.

[4] Xuemei Xie, Chenye Wang, Jiang Du. Full image recover for block-based compressive sensing [C]. IEEE International Conference on Multimedia and Expo (ICME), July 2018.

[5] Xuemei Xie, Yuxiang Wang, Guangming Shi, Chenye Wang, Jiang Du, Xiao Han, Adaptive measurement network for CS image reconstruction, CCCV 2017.

[6] Xuemei Xie, Zhifu Zhao, Chenye Wang, Jiang Du, Guangming Shi, Visualizing and Understanding of Learned Compressive Sensing with Residual Network, submitted to Neurocomputing.

[7] Zhifu Zhao, Xuemei Xie, Wan Liu, Chenye Wang, Jiang Du, Guangming Shi, A Hybrid-3D Convolutional Network for Video Compressive Sensing, submitted to AAAI 2018.

[8] Jiang Du, Xuemei Xie, Chenye Wang, Guangming Shi, Towards Real Compressive Sensing: From Compressive Sampling to Compressive Perception, submitted to IEEE Signal Processing Magazine.

 

3. Human Action Recognition

Action recognition includes predicting not only abnormal events like stealing, murdering and other criminals, but also the trends. Thus even we can see though the camera what people are thinking.

项目:智能视频监测及人员危害行为预警技术

Papers:

[1] Jianan Li, Xuemei Xie, Jianyu Chen, Yuhan Cao, Qingzhe Pan, Zhifu Zhao, Jiang Du, Guangming Shi. Guided Graph Convolutional Neural Network for Human Interaction Recognition, submitted to TPAMI 2018.

[2] Xuemei Xie, Wang Li, Jianan Li, Xun Xu. Local feature analysis for real-time action recognition, International Conference on Deep Learning Technology (ICDLT), 2018.

 

4. Object Detection

Research Interest: Small object detection, UAV imagery, Fine-grain objection detection, etc.

Papers:

[1] Xuemei Xie, Wenzhe Yang, Guimei Cao, Jianxiu Yang, Real-time Vehicle Detection from UAV Imagery, IEEE BigMM 2018.

[2] Xuemei Xie, Liao Quan and Lihua Ma, Gated Feature Pyramid Network for Object Detection, PRCV 2018.

[3] Guimei Cao, Xuemei Xie, Wenzhe Yang, Quan Liao, Guangming Shi, "Feature-Fused SSD: Fast Detection for Small Objects", ICGIP 2017.

 

5. DVS (Dynamic Vision Sensor)

                                                    

DVS is an event-based and frame-free camera which captures the changing pixels in video, and it generates the captured infomations in form of spikes. The figure below is accumulated from the events of a short period of time.

 

The first line of images are taken from a moving DVS camera on a driving car, while the second line of images are from fixed DVS camera set on the road. On the left is the original image, and on the right is denoised image using a "Two-layer KSVD" method.

          

                                Original                                                                        Denoised

                                                                  from moving camera

 

          

                                Original                                                                        Denoised

                                                                from fixed camera

 

Papers:

[1] Xuemei Xie, Shu Zhang, Jinjian Wu, Xun Xu, Jiang Du, Wang Li, Chenye Wang, Guangming Shi, Jianyu Chen, A Real-Time Rock-Paper-Scissor Hand Gesture Recognition System Based on FlowNet and Dynamic Vision Sensor, submitted to IEEE BigMM 2018, rejected.

[2] Xuemei Xie, Jiang Du, Guangming Shi, Jianxiu Yang, Wan Liu, Wang Li, DVS Image Noise Removal using K-SVD Method, ICGIP 2017.

[3] Xuemei Xie, Jiang Du, Guangming Shi, Hong Hu, Wang Li, An improved apporoach for Visualizing Dynamic Vision Sensor and its Video Denoising, ICVIP 2017.

[4] Xuemei Xie, Shu Zhang, Jiang Du, Xun Xu, Wang Li, Guangming Shi, Jianyu Chen, Live demonstration: a rock-paper-scissor gesture recognition system based on FlowNet and dynamic vision sensor, submitted to ICME 2018, rejected.

 

6. Visualization and pruning    

            

Papers:

Xuemei Xie, Xiao Han, Quan Liao, Guangming Shi, Visualization and Pruning of SSD with the base network VGG16, ICDLT 2017.

 

7. Intelligent transportation system & Real-time illigal parking system         

 

奖项:“TXI互联网+智能交通系统”在第十八届中国科协年会全国科技工作者创新创业大赛(陕西赛区)上获得银奖,2016年7月。

项目:智能交通卡口车类精准识别与统计方法及应用研究,2017年2月。

Papers:

[1] Xuemei Xie, Xun Xu, Lihua Ma, Pengfei Chen, Guangming Shi, "On the study of predictors in Single Shot Multibox Detector", ICVIP 2017.

[2] Xuemei Xie, Chenye Wang, Shu Chen, Guangming Shi, Zhifu Zhao, "Real-Time Illegal Parking Detection System Based on Deep Learning", ICDLT 2017

 

8. Neuroscience and Brain-inspired Computation

 

9. Research Activity

 

    

                        Research team    

 

Papers:

[1] Xuemei Xie, Zhang yazhong, Wu jinjian, Bag-of-Words Feature Representation for Blind Image Quality Assessment with Local Quantized Pattern, Neurocomputing, May 2017.

[2] Chao Wen, Xuemei Xie, Off-grid DOA estimation under nonuniform noise via variational sparse Bayesian learning, Signal Processing, May 2017.

版权声明:未经西安电子科技大学以及个人主页教师本人许可不得复制,转载传播本主页内容或用于任何商业用途,如需转载请征得主页所有者同意,并附上主页链接.
西安电子科技大学个人主页 站点地图 | | 谢雪梅的留言板 | 更新日期 :2018-10-30 | |
Copyright © 2011-2016 西安电子科技大学 版权所有 陕ICP备021284(05016169) Powered by 信息化建设处 & 电院网络中心