基本信息

韩振军  男  硕导/中科院青促会员
电子邮件:   hanzhj@ucas.ac.cn

实验室网址:http://vision.ucas.ac.cn

通信地址:   北京市怀柔区中国科学院大学学园二459室
邮政编码:   100049


I am currently recruiting Master students on the area of Image processing, Computer vision and Machine learning, please drop me an email with your detailed information if interested.


博士后及研究助理招聘:

方向:弱小目标检测与跟踪,多源信息融合,场景分割等

待遇:国科大待遇+课题组补助

欢迎申请!!!

部门/实验室:电子电气与通信工程学院

研究领域

智能视频监控

​遥感/航拍图像处理与分析


Latest:

Mar 2020 — We release the TinyBenchmark to help researchers develop better models and algorithms for tiny object detection (Link).

Feb 2020 — We (I am the General Chair) will organize the Real-world Recognition from Low-quality Inputs (RLQ) Workshop and the 1st Tiny Object Detection (TOD) Challenge with ECCV 2020 (Link).

Dec 2019 — I have one paper accepted by WACV 2020.

Sep 2019 — I have one paper accepted by TOMM.

Aug 2019 — I have one paper accepted by T-PAMI.

May 2019 — I have one paper accepted by PR.

Dec 2018 — I have one paper accepted by T-CSVT as the 1st author.

more+

招生信息

招生专业:

信号与信息处理、计算机应用技术、软件工程、数学


招生方向:

图像处理、计算机视觉、机器学习

教育背景

2009-09--2012-01   中国科学院研究生院   博士学位
2006-09--2009-07   中国科学院研究生院   硕士学位
2002-09--2006-07   天津大学   学士学位

工作经历

2013.06-至今  中国科学院大学   副教授
2011.12-至今  中国科学院大学   师资博士后 

教授课程

数字图像处理与分析
信号与系统
图像理解与应用
数字图像处理(电子与通信类)
开源硬件
人因工程学
数字图像处理

专利与奖励

奖励: 

1. 中国科学院青年创新促进会, 2019

2. 小波域图像表示与稀疏性目标感知, 二等奖, 部委级, 2015

3. 中国科学院院长奖, 2012


专利: 

1.  基于集群轨迹分类的集群场景智能监控方法及系统,发明专利授权ZL201510100197.0,2018

2.  无线视频监控系统及其图像传输方法和装置,发明专利授权,ZL201710028616.3,2019

3.  基于弱监督和深度响应重分配的X光图片违禁品定位方法,发明专利授权,ZL201811582841.2,2020

4.  基于分层传播和激活的弱监督X光图像违禁品检查方法,发明专利授权,ZL201811584403.X,2020

5. 一种基于尺度匹配的弱小人物目标检测方法,发明专利授权,ZL201910918836.2,2020

6.  一种基于无监督局部度量学习和重排序的行人再识别方法,发明专利授权,ZL201710677102.0, 2020

7. 一种基于选择性稀疏采样的细粒度图像分类方法,发明专利授权,ZL201910942790.8,2020

8. 一种基于分歧学习的弱监督目标分类和定位方法,发明专利授权,ZL201910942565.4,2020

9. 一种基于网络监督的目标分类和定位方法,发明专利申请,2019

10. 基于多源信息融合的弱小目标检测,发明专利申请,2020

11. 面向于弱小目标检测的深度卷积神经网络融合因子研究发明专利申请,2020

12. 基于精确尺度匹配的弱小人体目标检测发明专利申请,2020

13.  一种基于无监督深度孪生网络的视频去重方法,发明专利申请2020


软著:

1.  基于图割和区域对比度金字塔图像融合软件V1.0,2015

2.  基于尺度匹配的弱小航拍人体目标检测软件V1.0,2019 

3.  基于多源信息融合的弱小目标检测软件V1.02020

4.  基于精细尺度匹配的弱小目标检测软件V1.02020

5.  基于融合因子的弱小目标检测软件V1.02020


欢迎合作!!!

学术论文及兼职

部分期刊:

(1)Adaptive Discriminative Deep Correlation Filter for Visual Object Tracking,TCSVT,2019,第1作者 
(2)High performance person re-identification via a boosting ranking ensemble,Pattern Recognition,2019,通讯

(3)Spatial Preserved Graph Convolution Networks for Person Reidentification,Transactions on Multimedia Computing Communications and Applications(TOMM),2019,通讯

(4) Min-Entropy Latent Model for Weakly Supervised Object Detection,IEEE Trans. Pattern Anal. Mach. Intell.,2019,第3作者 

(5)CircleNet: Reciprocating Feature Adaptation for Robust Pedestrian Detection,IEEE Transactions on Intelligent Transportation Systems,2019, 第2作者

(6)A bottom-up summarization algorithm for videos in the wild,EURASIP J. Adv. Sig. Proc.,2019,第4作者

(7)Beyond Group: Multiple Person Tracking via Minimal Topology-Energy-Variation, IEEE Trans. Image Processing,2017,第5作者
(8)Real-Time Multi-pedestrian Tracking in Traffic Scenes via an RGB-D-Based Layered Graph Model, IEEE Trans. Intelligent Transportation Systems, 2015,第2作者

(9) Visual Object Tracking via Sample-Based Adaptive Sparse Representation,Pattern Recognition,2011,第1作者 
(10) Combined feature evaluation for adaptive visual object tracking,Computer Vision and Image Understanding,2011,第1作者
(11) Human Detection In Images Via Piecewise Linear Support Vector,IEEE Transactions on Image Processing,2012,第2作者 
(12) Visual Abnormal Behavior Detection Based on Trajectory Sparse Reconstruction Analysis,Neurocomputing,2012,第2作者 

 

部分会议:

(1)Scale Match for Tiny Person Detection,WACV,2020,通讯

(2)Feature Hourglass Network for Skeleton Detection,CVPR workshop,2019,通讯(竞赛排名第二)

(3)Min-Entropy Latent Model for Weakly Supervised Object Detection,CVPR,2018,第4作者

(4)A scalable convolutional neural network for task-specified scenarios via knowledge distillation,ICASSP,2017,第4作者

(5)Unsupervised person re-identification via re-ranking enhanced sample-specific metric learning,ICIP,2017,通讯

(6)Person re-identification via adaboost ranking ensemble,ICIP,2016,第2作者

(7)Locality-Constrained Sparse Reconstruction for Trajectory Classification,ICPR,2014,第2作者

(8)Depth Structure Association for RGB-D Multi-target Tracking,ICPR,2014,第2作者

(9) Robust Visual Object Tracking via Sparse Representation and Reconstruction,International Conference on Computer Analysis of Images and Patterns,2013,第1作者
(10) A Fast Object Tracking Approach Based on Sparse Representation,International Conference on Image Processing,2011,第1作者 
(11) Online Feature Evaluation for Object Tracking Using Kalman Filter,International Conference on Pattern Recognition,2008,第1作者 
(12) Feature Evaluation by Particle Filter for Adaptive Object Tracking,International Conference on Visual Communications and Image Processing ,2008,第1作者
(13) Real-time Multi-pedestrian Tracking Based on Vision and Depth information fusion,Proc. of Pacific-Rim Conference on Multimedia, 2013,第2作者
(14) A Belief Based Correlated Topic Model for Semantic Region Analysis in Far-Field Video Surveillance Systems,The Pacific-Rim Conference on Multimedia,2013,第四作者 

(15) Robust Weld Line Detection with Cross Structured Light and Hidden Markov Model,Proc. of 2012 IEEE International Conference on Mechatronics and Automation,2012,第4作者 
(16) Visual Object Tracking via One-Class SVM,ACCV2010 Workshop on Visual Surveillance,2010,第2作者  


部分学术兼职:

(1)Associate Editor of Journal of Electronic Imaging

(2)Guest Editor of Special issue on Semantic representations for social behavior analysis in video surveillance systems, Signal, Image and Video Processing
(3) Session Chair and Meta-Reviewer of special issue: Multimedia Information Processing and Understanding for Intelligent Surveillance, 4th International Conference on Internet Multimedia Computing and Service


发表著作:

(1) 视觉目标检测与跟踪,科学出版社,2016年