Weihong Deng





Associate Professor

Pattern Recognition and Intelligent System Laboratory
(PRIS Lab), Beijing University of Posts and Telecommunications, Beijing 100876, China

E-Mail: myname@bupt.edu.cn, myname="whdeng"




Tutorial talk at CVPR 2015

Tutorial talk at FG 2015

Tutorial talk at ACCV 2014

Co-organizer of ACCV 2014 Workshop

Tutorial talk at ICME 2014












Research Interests

With ten years’ research and development experience on computer vision and pattern recognition, with speical foucs on

Research Projects

Transform-Invariant PCA PAMI 2014
TIPCA technique learn the coding bases invariant to the transformation of the training images, leading to improved alignment, coding, and recognition performance

Extended Sparse Repesenation PAMI 2012, CVPR2013
Sparse Representation based Classification (SRC) is a face-recognition breakthrough in recent years, which has successfully addressed the recognition problem with sufficient training images of each gallery subject. In this project, we extend SRC to applications where there are very few, or even a single, training images per subject.

Biologically Inspired Face Recognition PR2010, SCIENCE 2008
Face recognition technology is of great significance for applications involving national security and crime prevention. Despite enormous progress in this field, machine-based system is still far from the goal of matching the versatility and reliability of human face recognition. In this project, we show that a simple system designed by emulating biological strategies of human visual system can largely surpass the state-of-the-art performance on uncontrolled face recognition.

"Uniform Inter-Class Distance" Criterion for Feature Extraction PR2010,PR2012,PR2014
Current face recognition techniques rely heavily on the large size and representativeness of the training sets, and most methods suffer degraded performance or fail to work if there is only one training sample per person available. This so-called 'one sample problem' is a challenging issue in face recognition. In this project, we propose a novel feature extraction method named uniform pursuit to address the one sample problem.

High-dimensional Pattern Analysis PAMI2008, SIGIR2007, PR2012
This project aims to conduct indeep theoretical analysis on selected problems of the high-dimensional data analysis

1) Justify the close relationship, or equivalence, among the well-known feature extraction algorithms, such as UDP and LPP.

2) Propose a locality discriminating indexing (LDI) algorithm for document classifcation, which seeks for a projection whichbest preserves the within-class local structures while suppresses the between-class overlap.

3) Analyze the small sample size problem of ICA.


Research Recommendations


Real-world Affective Face

Cross-Age LFW

Similar-Looking LFW

Publications (Google Scholar Profile)

Journal Papers

Conference Papers


Professional Services

Reviewers for


Kinship Verification & Metric Learning Data Set by Dr. Jiwen Lu
Methodology of CVPR Research by Prof. Shiguang Shan (In Chinese)