1. 多媒体技术，计算机视觉、机器学习；

2. 交通大数据、智能交通、数据驱动的交通规划；

个人简介

银河集团9873.cσm人工智能与自动化学院，“多媒体与智能软件技术”北京市重点实验室研究人员。

中国计算机学会(China Computer Federatio, CCF）会员，CCF多媒体专委会委员，CCF人工智能专委会通讯委员，电气和电子工程师协会(Institute of Electrical and Electronics Engineers, IEEE)会员。

DBLP:https://dblp.uni-trier.de/pers/hd/p/Pang:Junbiao
       ORCiD: https://orcid.org/0000-0001-8153-7229
   更多最新信息可参考实验室主页: https://learningandvision.github.io

教育简历

2005/9～2011/7，中国科学院计算技术研究所，计算机应用技术，博士

2002/9～2004/7，哈尔滨工业大学，计算机科学与技术，硕士

1998.9～2002.7，哈尔滨工业大学，计算流体力学，学士

工作履历

2016/7～至今，北京工业大学，银河集团9873.cσm，副教授（博士生/硕士生导师）

2013/12～2016/12，北京工业大学，城市交通学院，讲师/副教授（硕士生导师）

2011/7～2013/12，北京工业大学，计算机学院，讲师

2010/9～2011/7，NEC（日电）中国研究院，视觉组，访问学者

2004/7～2005/7，华为技术有限公司，固网部门，工程师

课程教学

本科生教学：设计模式、Python语言

研究生教学：模式识别、机器学习

科研项目

作为项目负责人和学术骨干承担了多项国家和省市级项目的研究工作，主要研究项目包括：

1. 国家自然科学基金面上项目，61472387, 注意选择引导的人体运动分析和识别，2015/01-2018/12, 80万，已结题，参加

2. 国家自然科学基金重点项目，61332016，面向网络事件的跨平台网络异质媒体语义协同与挖掘，2014/01-2018/12，300万，已结题，参加

3. 国家自然科学基金面上项目，61672069, 基于异质数据的城市公交到站时间预测研究，2017/01-2020/12, 63万，已结题，主持

4. 北京市自然科学基金面上项目，4132010，真实环境下针对姿态多变和遮掩问题的物体检测方法研究，2013/01-2015/12，14万，已结题，主持

5. 国家自然科学基金（青年基金项目），61202234，弱标注下基于主动学习的物体检测适用问题研究，2013/01-2015/12，24万，已结题，主持

6. 国家自然科学基金面上项目，61872333, 弱监督下的视频动作检测，2019/01-2022/12, 65万，已结题，参加

7. 若干企业和事业单位横向项目

荣誉和获奖

视觉边缘计算装备与关键技术中国仪器仪表学会技术发明奖（二等奖），排名第二，2022

代表性研究成果

近年来，在国内外重要学术刊物和会议上发表文章30多篇，并申请多项发明专利。代表性研究成果包括：

[1] Junbiao Pang ; Jing Huang ; Yong Du ; Haitao Yu ; Qingming Huang ; Baocai Yin：Learning to Predict Bus Arrival Time From Heterogeneous Measurements via Recurrent Neural Network,　IEEE Transactions on Intelligent Transportation Systems (SCI, JCR Q1) ( Early Access )

[2] Junbiao Pang ; Anjing Hu ; Qingming Huang ; Qi Tian ; Baocai Yin: Increasing Interpretation of Web Topic Detection via Prototype Learning From Sparse Poisson Deconvolution,IEEE Transactions on Cybernetics (SCI, JCR Q1) ( Early Access )

[3] Junbiao Pang ; Fei Tao ; Qingming Huang ; Qi Tian ; Baocai Yin: Two Birds With One Stone: A Coupled Poisson Deconvolution for Detecting and Describing Topics From Multimodal Web Data,IEEE Transactions on Neural Networks and Learning Systems(SCI, JCR Q1) ( Early Access )

[4] Jinzhong Lin, Junbiao Pang, Li Su, Yugui Liu, Qingming Huang:Accelerating Topic Detection on Web for a Large-Scale Data Set via Stochastic Poisson Deconvolution. MMM (1) 2019: 590-602

[5] Junbiao Pang, Fei Tao, Liang Li, Qingming Huang, Baocai Yin, Qi Tian: A two-step approach to describing web topics via probable keywords and prototype images from background-removed similarities. Neurocomputing 275: 478-487 (2018) (SCI, JCR Q2)

[6] Junbiao Pang, Jing Huang, Xue Yang, Zuyun Wang, Haitao Yu, Qingming Huang, Baocai Yin: Discovering Fine-Grained Spatial Pattern From Taxi Trips: Where Point Process Meets Matrix Decomposition and Factorization. IEEE Trans. Intelligent Transportation Systems 19(10): 3208-3219 (2018) (SCI, JCR Q1)

主要论文论著

[1] Junbiao Pang, Jing Huang, Lei Qin, Weigang Zhang, Laiyun Qing, Qingming Huang, Baocai Yin:Rotative maximal pattern: A local coloring descriptor for object classification and recognition. Inf. Sci. 405: 190-206 (2017) (SCI, JCR Q1)

[2] Junbiao Pang, Jing Huang, Weigang Zhang, Qingming Huang, Baocai Yin:Justify role of Similarity Diffusion Process in cross-media topic ranking: an empirical evaluation. Multimedia Tools Appl. 76(23): 25145-25157 (2017) (SCI, JCR Q3)

[3] Junbiao Pang, Fei Tao, Chunjie Zhang, Weigang Zhang, Qingming Huang, Baocai Yin:Robust Latent Poisson Deconvolution From Multiple Features for Web Topic Detection. IEEE Trans. Multimedia 18(12): 2482-2493 (2016) (SCI, JCR Q1)

[4] Junbiao Pang, Huihuang Lin, Li Su, Chunjie Zhang, Weigang Zhang, Lijuan Duan, Qingming Huang, Baocai Yin:Accelerate convolutional neural networks for binary classification via cascading cost-sensitive feature. ICIP 2016: 1037-1041 (CCF B)

[5]Junbiao Pang, Chunjie Zhang, Lei Qin, Weigang Zhang, Laiyun Qing, Qingming Huang, Baocai Yin:Online dictionary learning for Local Coordinate Coding with Locality Coding Adaptors. Neurocomputing 157: 61-69 (2015) (SCI, JCR Q2)

[6]Junbiao Pang, Lei Qin, Chunjie Zhang, Weigang Zhang, Qingming Huang, Baocai Yin:Local Laplacian Coding From Theoretical Analysis of Local Coding Schemes for Locally Linear Classification. IEEE Trans. Cybernetics 45(12): 2937-2947 (2015) (SCI, JCR Q1)

[7] Chunjie Zhang, Jian Cheng, Jing Liu, Junbiao Pang, Qingming Huang, Qi Tian:Beyond Explicit Codebook Generation: Visual Representation Using Implicitly Transferred Codebooks. IEEE Trans. Image Processing 24(12): 5777-5788 (2015) (SCI, JCR Q1)

[8] Junbiao Pang, Fei Jia, Chunjie Zhang, Weigang Zhang, Qingming Huang, Baocai Yin:Unsupervised Web Topic Detection Using A Ranked Clustering-Like Pattern Across Similarity Cascades. IEEE Trans. Multimedia 17(6): 843-853 (2015) (SCI, JCR Q1)

[9] Junbiao Pang, Qingming Huang, Baocai Yin, Lei Qin, Dan Wang: Theoretical analysis of learning local anchors for classification. ICPR 2012: 1803-1806 (CCF B)

[10] Qingyuan Wang, Junbiao Pang, Guoyi Liu, Lei Qin, Qingming Huang, Shuqiang Jiang:Color Maximal-Dissimilarity Pattern for pedestrian detection. ICPR 2012: 1952-1955 (CCF B)

[11] Junbiao Pang, Qingming Huang, Shuicheng Yan, Shuqiang Jiang, Lei Qin: Transferring Boosted Detectors Towards Viewpoint and Scene Adaptiveness. IEEE Trans. Image Processing 20(5): 1388-1400 (2011) (SCI, JCR Q1)

[12] Junbiao Pang, Qingming Huang, Shuqiang Jiang:Multiple Instance Boost Using Graph Embedding Based Decision Stump for Pedestrian Detection. ECCV (4) 2008: 541-552