Basic Information


Deyu Meng (孟德宇

Associate Professor


Institute for Information and System Sciences

Faculty of Mathematics and Statistics

Xi’an Jiaotong University 

Xi’an, Shaan’xi Province, P. R. China, 710049


Fax(传真):   +86 29 8266 8559 
Tel(电话):  +86 29 8266 3153

XJTU Homepage(西安交通大学主页):

CMU Homepage(卡内基梅隆大学主页):

Email(邮箱): dymeng at mail dot xjtu dot edu dot cn




Welcome to experience the performance of our newly published method on MSI denoising :

Yi Peng, Deyu Meng, Zongben Xu, Chenqiang Gao, Yi Yang, Biao Zhang. Decomposable Nonlocal Tensor Dictionary Learning for Multispectral Image DenoisingSupplementary Material, CVPR, 2014. Matlab code.

(The provided code has a small bug. We have just fixed it. Sorry for the previous users who encountered this problem.)

Our robust PCA method can well fit more complex noise beyond conventional Gaussian, Laplacian, Sparse noise or any combinations of them:

Qian Zhao, Deyu Meng, Zongben Xu, Wangmeng Zuo, Lei Zhang. Robust principal component analysis with complex noiseSupplementary MaterialICML, 2014, Matlab code

Our proposed FastMMD method decreases the time complexity of MMD calculation from conventional O(N^2d) to O(Nlog(d)):

Ji Zhao, Deyu Meng. FastMMD: Ensemble of Circular Discrepancy for Efficient Two-Sample Test, arXiv:1405.2664, 2014, Matlab Code

The code and slides of our ICMR best-paper-runner-up work are now available:

 Lu Jiang, Wei Tong, Deyu Meng, Alexander G. Hauptmann.Towards Efficient Learning of Optimal Spatial Bag-of-Words Representations. ICMR, 2014. slides, Please download our code in JS Tiling webpage

The corrected proof to the global optimum of weighted nuclear norm minimization:

Qi Xie, Deyu Meng, Shuhang Gu, Lei Zhang, Wangmeng Zuo, Xiangchu Feng, Zongben Xu. On the optimal solution of weighted nuclear norm minimization, arXiv:1405.6012v1, 2014.

Our new method can effectively parse human motion in unconstrained Internet videos WITHOUT need to label any videos for training:

Haoquan Shen, Shoou-I Yu, Yi Yang, Deyu Meng, Alexander Hauptmann. Unsupervised Video Adaptation for Parsing Human Motion. ECCV, 2014. Demo